Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: CastError
Message: Couldn't cast
ok: bool
input: string
record_count: int64
clean_count: int64
train_count: int64
val_count: int64
warnings: list<item: null>
child 0, item: null
outputs: struct<train: string, val: string>
child 0, train: string
child 1, val: string
context: string
source: string
tags: list<item: string>
child 0, item: string
output: string
instruction: string
confidence: double
to
{'instruction': Value('string'), 'input': Value('string'), 'context': Value('string'), 'output': Value('string'), 'tags': List(Value('string')), 'source': Value('string'), 'confidence': Value('float64')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1872, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 260, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 120, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2272, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
ok: bool
input: string
record_count: int64
clean_count: int64
train_count: int64
val_count: int64
warnings: list<item: null>
child 0, item: null
outputs: struct<train: string, val: string>
child 0, train: string
child 1, val: string
context: string
source: string
tags: list<item: string>
child 0, item: string
output: string
instruction: string
confidence: double
to
{'instruction': Value('string'), 'input': Value('string'), 'context': Value('string'), 'output': Value('string'), 'tags': List(Value('string')), 'source': Value('string'), 'confidence': Value('float64')}
because column names don't match
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1347, in compute_config_parquet_and_info_response
parquet_operations = convert_to_parquet(builder)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
builder.download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 884, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 947, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1739, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1922, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
instruction string | input string | context string | output string | tags list | source string | confidence float64 |
|---|---|---|---|---|---|---|
Assess this network session and recommend defensive next steps. | session_id=SID_00635, protocol=TCP, packet_size=297, login_attempts=2, failed_logins=2, session_duration=933.2s, encryption=AES, ip_reputation_score=0.10, browser=Chrome, unusual_time_access=yes | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=297, login_attempts=2, failed_logins=2, unusual_time_access=yes, encryption=AES, reputation_score=0.10, browser=Chrome, session_duration=933.2s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) verify whether the access time aligns with approved maintenance or travel activity. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Turn this cyber incident trend into a defensive planning brief. | country=Australia, year=2023, attack_type=Man-in-the-Middle, target_industry=Banking, vulnerability=Zero-day, defense_used=Firewall, attack_source=Unknown, financial_loss_musd=57.34, affected_users=471743, resolution_hours=29 | Provide risk prioritization, hardening guidance, and monitoring recommendations for a blue-team audience. | Risk summary: Man-in-the-Middle against the Banking sector in Australia shows a high defensive planning priority. Observed weaknesses center on Zero-day, with reported source attribution of Unknown; existing control listed was Firewall. Planning actions: 1) review whether your environment has the same exposure class and close the weakness through patching, stronger identity controls, or phishing resistance as appropriate, 2) ensure detections cover early indicators for man-in-the-middle activity and escalation paths in the banking workflow, 3) prepare business-impact communications because reported impact reached $57.34M and 471743 affected users, 4) shorten containment and recovery playbooks if current resolution times approach 29 hours. | [
"threat-intel",
"risk-planning",
"man-in-the-middle",
"banking"
] | atharvasoundankar/global-cybersecurity-threats-2015-2024 | 0.72 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=prozadc valiuam xanwax patxil prowzac zolxoft celerbrex viowxx ciaxlis prophecia viacgra viaqgra st ambcien zybgan no perszcriptdion is requtired disucreet ovternizght shipmping to your dotor stop ovetrpawying on yoxur mezds 100 % monney bajck guaroantee on all purcthases the old figure stirred me as it had never done before and seemed a good omen never seen any blood so bright . when he lay down and turned margie... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=TimH:>I was watching that last night and I couldn't help wondering if one >of the GIs was Dexter Fletcher out of the Crystal Maze. Yup, was him out of Press Gang>Who once snogged Shauna Lowery, if anyones heard of her. Did you have to kneel down? DaveTo unsubscribe from this group, send an email to: [redacted-email] Your use of Yahoo! Groups is subject to [redacted-url] | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Assess this network session and recommend defensive next steps. | session_id=SID_06692, protocol=ICMP, packet_size=359, login_attempts=3, failed_logins=1, session_duration=528.1s, encryption=AES, ip_reputation_score=0.36, browser=Firefox, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=ICMP, packet_size=359, login_attempts=3, failed_logins=1, unusual_time_access=no, encryption=AES, reputation_score=0.36, browser=Firefox, session_duration=528.1s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"icmp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Assess this network session and recommend defensive next steps. | session_id=SID_01130, protocol=TCP, packet_size=565, login_attempts=6, failed_logins=0, session_duration=199.1s, encryption=AES, ip_reputation_score=0.33, browser=Unknown, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=TCP, packet_size=565, login_attempts=6, failed_logins=0, unusual_time_access=no, encryption=AES, reputation_score=0.33, browser=Unknown, session_duration=199.1s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=rolex watches - for as low as $ 65 ! buy a quality watch today and receive free shipping and handling ! over one million sold ! inventory is getting low ! 100 % guaranteed of lowest prices and highest quality ! would you be interrested in wearingarolex ora tag heuer ? rolex - menswatches rolex - ladies watches over one thousand to choose from here we love to please our customers ! i am 100 % sure you will be pleas... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Quantum Token Spoofing via Zigbee Fake Cluster Injection, category=Quantum Cryptography & Post-Quantum Threats, attack_type=Wireless - Zigbee Cluster Emulation, scenario=Attacker injects fake Zigbee cluster attributes matching quantum token handshake data, target_type=Zigbee Quantum Security Devices, vulnerability=Insecure cluster field validation during token handshake, mitre_technique=T1071.002 | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Quantum Token Spoofing via Zigbee Fake Cluster Injection is an Wireless - Zigbee Cluster Emulation scenario in Quantum Cryptography & Post-Quantum Threats affecting Zigbee Quantum Security Devices. Primary weakness: Insecure cluster field validation during token handshake. MITRE mapping: T1071.002. Expected impact if exploited: Bypassed authentication via field emulation. Detection guidance: Monitor Zigbee cluster field consistency. Hardening guidance: Harden attribute parsing and entropy-based challenge. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"wireless - zigbee cluster emu...",
"quantum cryptography & post-q...",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=Hello I am your hot lil horny toy. I am the one you dream About, I am a very open minded person, Love to talk about and any subject. Fantasy is my way of life, Ultimate in sex play. Ummmmmmmmmmmmmm I am Wet and ready for you. It is not your looks but your imagination that matters most, With My sexy voice I can make your dream come true... Hurry Up! call me let me Cummmmm for you.......................... TOLL-FREE... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Assess this network session and recommend defensive next steps. | session_id=SID_02418, protocol=TCP, packet_size=568, login_attempts=3, failed_logins=3, session_duration=931.3s, encryption=AES, ip_reputation_score=0.47, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=TCP, packet_size=568, login_attempts=3, failed_logins=3, unusual_time_access=no, encryption=AES, reputation_score=0.47, browser=Chrome, session_duration=931.3s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=wellesley list geynille : here is the list of the women who attended the presentation at wellesley along with their e - mail addresses . laura | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Assess this network session and recommend defensive next steps. | session_id=SID_04074, protocol=UDP, packet_size=471, login_attempts=5, failed_logins=1, session_duration=434.7s, encryption=AES, ip_reputation_score=0.29, browser=Edge, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: low. Evidence: protocol=UDP, packet_size=471, login_attempts=5, failed_logins=1, unusual_time_access=no, encryption=AES, reputation_score=0.29, browser=Edge, session_duration=434.7s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"udp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Turn this cyber incident trend into a defensive planning brief. | country=Brazil, year=2022, attack_type=DDoS, target_industry=Telecommunications, vulnerability=Unpatched Software, defense_used=Firewall, attack_source=Hacker Group, financial_loss_musd=79.50, affected_users=789067, resolution_hours=63 | Provide risk prioritization, hardening guidance, and monitoring recommendations for a blue-team audience. | Risk summary: DDoS against the Telecommunications sector in Brazil shows a urgent defensive planning priority. Observed weaknesses center on Unpatched Software, with reported source attribution of Hacker Group; existing control listed was Firewall. Planning actions: 1) review whether your environment has the same exposure class and close the weakness through patching, stronger identity controls, or phishing resistance as appropriate, 2) ensure detections cover early indicators for ddos activity and escalation paths in the telecommunications workflow, 3) prepare business-impact communications because reported impact reached $79.50M and 789067 affected users, 4) shorten containment and recovery playbooks if current resolution times approach 63 hours. | [
"threat-intel",
"risk-planning",
"ddos",
"telecommunications"
] | atharvasoundankar/global-cybersecurity-threats-2015-2024 | 0.72 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Malware, severity=Medium, protocol=ICMP, traffic_type=HTTP, packet_length=1464, anomaly_score=19.30, attack_signature=Known Pattern A, malware_indicators=IoC Detected, alerts=Alert Triggered, action_taken=Ignored, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Malware telemetry with elevated event that merits analyst review. Evidence: severity=Medium, protocol=ICMP, traffic_type=HTTP, packet_length=1464, anomaly_score=19.30, signature=Known Pattern A, indicators=IoC Detected, alerts=Alert Triggered, current_action=Ignored, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"malware",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Assess this network session and recommend defensive next steps. | session_id=SID_02201, protocol=TCP, packet_size=860, login_attempts=8, failed_logins=2, session_duration=128.2s, encryption=DES, ip_reputation_score=0.53, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=TCP, packet_size=860, login_attempts=8, failed_logins=2, unusual_time_access=no, encryption=DES, reputation_score=0.53, browser=Chrome, session_duration=128.2s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=here ' s a hot piay in motion " stock watch alert " this morning are wysak petroleum ( wysk ) , key energy services , inc . ( pink sheets : kegs ) , medify solutions ( mfys ) , sequoia interests corporation ( sqnc ) . wysak petroleum ( wysk ) current price : o . 17 wysak petroleum announces the signing of a letter of intent with the european commission baltic renewabie energy centre ( ec brec ) to assist wysak pet... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=GPS Jamming of Port Container Tracking Systems, category=Satellite & Space Infrastructure Security, attack_type=GPS Jamming, scenario=Disrupting GPS signals on container tracking devices causing inventory management issues at ports., target_type=Port Infrastructure, vulnerability=No anti-jamming countermeasures, mitre_technique=MITRE T1499 (Endpoint DoS) | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: GPS Jamming of Port Container Tracking Systems is an GPS Jamming scenario in Satellite & Space Infrastructure Security affecting Port Infrastructure. Primary weakness: No anti-jamming countermeasures. MITRE mapping: MITRE T1499 (Endpoint DoS). Expected impact if exploited: Inventory loss, shipment delays, logistical chaos. Detection guidance: Tracking data anomaly detection, GPS signal quality monitoring. Hardening guidance: Use encrypted tracking signals and anti-jamming technology. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"gps jamming",
"satellite & space infrastruct...",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=FTP Banner Grabbing, category=AI Agents & LLM Exploits, attack_type=Reconnaissance / Fingerprinting, scenario=An attacker or penetration tester connects to an FTP server (File Transfer Protocol) to retrieve banner information like the FTP server software name, version, and operating system. This info helps identify vulnerabilities in outdated software. The banner is usually revealed immediately when you connect, even without logging in., target_type=FTP Servers (VSFTPD, ProFTPD, FileZilla Server, etc.), vulnerability=Information disclosure via service greeting banners, mitre_technique=T1592.004 – Gather Victim Host Information: Server | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: FTP Banner Grabbing is an Reconnaissance / Fingerprinting scenario in AI Agents & LLM Exploits affecting FTP Servers (VSFTPD, ProFTPD, FileZilla Server, etc.). Primary weakness: Information disclosure via service greeting banners. MITRE mapping: T1592.004 – Gather Victim Host Information: Server. Expected impact if exploited: - Identifies FTP server version- Enables targeted vulnerability scanning- Risk of full exploit via known CVEs. Detection guidance: - Monitor port 21 access logs- Use IDS/IPS to track FTP banner grabbing attempts. Hardening guidance: - Disable banner (in vsftpd.conf: ftpd_banner=)- Update FTP software regularly- Restrict anonymous access and whitelist IPs. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"reconnaissance / fingerprinting",
"ai agents & llm exploits",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Assess this network session and recommend defensive next steps. | session_id=SID_00388, protocol=ICMP, packet_size=875, login_attempts=4, failed_logins=3, session_duration=1016.0s, encryption=AES, ip_reputation_score=0.13, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=ICMP, packet_size=875, login_attempts=4, failed_logins=3, unusual_time_access=no, encryption=AES, reputation_score=0.13, browser=Chrome, session_duration=1016.0s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"icmp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=UART Access to Medical Infusion Pump, category=Physical / Hardware Attacks, attack_type=Hardware Interface Exploitation, scenario=Direct access to infusion pump console via UART to modify dosage logs., target_type=Medical Device, vulnerability=Unauthenticated serial CLI, mitre_technique=T0887 | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: UART Access to Medical Infusion Pump is an Hardware Interface Exploitation scenario in Physical / Hardware Attacks affecting Medical Device. Primary weakness: Unauthenticated serial CLI. MITRE mapping: T0887. Expected impact if exploited: Patient harm, data fraud. Detection guidance: Medical audit trail mismatch. Hardening guidance: Disable UART, use logging controller. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"hardware interface exploitation",
"physical / hardware attacks",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=High, protocol=ICMP, traffic_type=DNS, packet_length=1082, anomaly_score=68.79, attack_signature=Known Pattern A, malware_indicators=IoC Detected, alerts=, action_taken=Logged, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with high-priority incident candidate. Evidence: severity=High, protocol=ICMP, traffic_type=DNS, packet_length=1082, anomaly_score=68.79, signature=Known Pattern A, indicators=IoC Detected, alerts=, current_action=Logged, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=more energy ! body bgcolor = black table bgcolor = white cellpadding = 10 align = centertrtd valign = top width = 117 align = centertable border = 1 trtd align = center width = 115 a href = http : / / medsforyou . info / hgh / index . php ? pid = eph 9106 img brfont size = 4 face = arialtake yourbrlife back . . . brbra href = http : / / medsforyou . info / hgh / index . php ? pid = eph 9106 font size = 5 color = r... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Assess this network session and recommend defensive next steps. | session_id=SID_09447, protocol=TCP, packet_size=339, login_attempts=4, failed_logins=0, session_duration=1356.5s, encryption=DES, ip_reputation_score=0.38, browser=Chrome, unusual_time_access=yes | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=339, login_attempts=4, failed_logins=0, unusual_time_access=yes, encryption=DES, reputation_score=0.38, browser=Chrome, session_duration=1356.5s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=DDoS, severity=Low, protocol=UDP, traffic_type=DNS, packet_length=222, anomaly_score=24.62, attack_signature=Known Pattern B, malware_indicators=, alerts=Alert Triggered, action_taken=Logged, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: DDoS telemetry with lower-severity signal that still needs validation. Evidence: severity=Low, protocol=UDP, traffic_type=DNS, packet_length=222, anomaly_score=24.62, signature=Known Pattern B, indicators=, alerts=Alert Triggered, current_action=Logged, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"ddos",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Assess this network session and recommend defensive next steps. | session_id=SID_09441, protocol=TCP, packet_size=148, login_attempts=1, failed_logins=2, session_duration=443.2s, encryption=AES, ip_reputation_score=0.31, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=148, login_attempts=1, failed_logins=2, unusual_time_access=no, encryption=AES, reputation_score=0.31, browser=Chrome, session_duration=443.2s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=Medium, protocol=ICMP, traffic_type=HTTP, packet_length=96, anomaly_score=36.53, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=, action_taken=Blocked, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with elevated event that merits analyst review. Evidence: severity=Medium, protocol=ICMP, traffic_type=HTTP, packet_length=96, anomaly_score=36.53, signature=Known Pattern B, indicators=IoC Detected, alerts=, current_action=Blocked, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=[redacted-url] changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |WONTFIX------- Additional Comments From [redacted-email] 2002-10-03 14:16 ------- I think it's ok to close this one? Comment is a valid, albeit seldom used RFC 2822 header... ------- You are receiving this mail because: ------- You are the assignee for the bug,... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Alert on Azure Resource Deletion Spree, category=Blue Team, attack_type=Mass Deletion, scenario=Malicious actor deletes multiple resources across subscriptions, target_type=Azure, vulnerability=Destructive activity, mitre_technique=T1485 | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Alert on Azure Resource Deletion Spree is an Mass Deletion scenario in Blue Team affecting Azure. Primary weakness: Destructive activity. MITRE mapping: T1485. Expected impact if exploited: Resource loss. Detection guidance: Delete op correlation. Hardening guidance: Lock + IR + Recover. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"mass deletion",
"blue team",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Assess this network session and recommend defensive next steps. | session_id=SID_01404, protocol=UDP, packet_size=643, login_attempts=4, failed_logins=0, session_duration=374.7s, encryption=DES, ip_reputation_score=0.05, browser=Firefox, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=UDP, packet_size=643, login_attempts=4, failed_logins=0, unusual_time_access=no, encryption=DES, reputation_score=0.05, browser=Firefox, session_duration=374.7s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"udp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Ransomware via Fake PDF Invoice, category=Malware & Threat, attack_type=Ransomware Simulation, scenario=Attacker sends a fake invoice PDF that triggers a download via embedded link., target_type=Windows Desktop, vulnerability=User clicks embedded link, mitre_technique=T1204.001, T1059 | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Ransomware via Fake PDF Invoice is an Ransomware Simulation scenario in Malware & Threat affecting Windows Desktop. Primary weakness: User clicks embedded link. MITRE mapping: T1204.001, T1059. Expected impact if exploited: Data loss, deceptive invoice attack. Detection guidance: Email link logs, system file rename. Hardening guidance: Email filtering, PDF sanitization. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"ransomware simulation",
"malware & threat",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=re : possible hire ? kathy , thanks for your message . we are very interested in talking to keith . please , give him my phone number / e - mail address and he can contact me directly . vince " kathy ensor " on 04 / 18 / 2000 11 : 32 : 27 am to : cc : subject : possible hire ? dear vince , a faculty member in our department , namely dr . keith baggerly , is interested in pursuing other employment . he is currently... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Assess this network session and recommend defensive next steps. | session_id=SID_07008, protocol=TCP, packet_size=770, login_attempts=4, failed_logins=1, session_duration=1067.1s, encryption=AES, ip_reputation_score=0.18, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=770, login_attempts=4, failed_logins=1, unusual_time_access=no, encryption=AES, reputation_score=0.18, browser=Chrome, session_duration=1067.1s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=ICS Device Driver Installer Trojan, category=SCADA/ICS, attack_type=Supply Chain / Driver Trojan, scenario=A custom driver installer for a SCADA field device is trojanized and bundled with rootkit functionality., target_type=Engineering Workstation, vulnerability=Driver integrity not validated, mitre_technique=T1055 + T1543.003 | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: ICS Device Driver Installer Trojan is an Supply Chain / Driver Trojan scenario in SCADA/ICS affecting Engineering Workstation. Primary weakness: Driver integrity not validated. MITRE mapping: T1055 + T1543.003. Expected impact if exploited: Hidden persistent access into ICS zone. Detection guidance: Detect hidden services, DLL injection behavior. Hardening guidance: Only use signed drivers; EDR enforcement. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"supply chain / driver trojan",
"scada/ics",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=request submitted : access request for steve . moen @ enron . com you have received this email because you are listed as a security approver . please click approval to review and act upon this request . request id : 000000000007593 request create date : 11 / 16 / 00 11 : 21 : 54 am requested for : steve . moen @ enron . com resource name : eol us backoffice stack manager resource type : applications | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Malware, severity=Low, protocol=UDP, traffic_type=FTP, packet_length=769, anomaly_score=85.81, attack_signature=Known Pattern A, malware_indicators=, alerts=, action_taken=Ignored, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Malware telemetry with high-priority incident candidate. Evidence: severity=Low, protocol=UDP, traffic_type=FTP, packet_length=769, anomaly_score=85.81, signature=Known Pattern A, indicators=, alerts=, current_action=Ignored, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"malware",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=real products for real people . b dont waste your time at the doctors office ! rx medications delivered right to your door in 24 hours ! pay less for your drugs get more for your $ $ $ ! join the millions of people who are tired of the hassle with the insurance companies and doctors ! we carry all of the well - known drugs available and most of the unknown as well . we currently have specials on the following item... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Smart Doorbell - XSS via Video Caption Field, category=IoT / Embedded Devices, attack_type=XSS, scenario=The doorbell allows adding captions to saved visitor videos. The caption field is vulnerable to stored XSS., target_type=Smart Doorbell, vulnerability=Stored XSS, mitre_technique=T1059.007 | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Smart Doorbell - XSS via Video Caption Field is an XSS scenario in IoT / Embedded Devices affecting Smart Doorbell. Primary weakness: Stored XSS. MITRE mapping: T1059.007. Expected impact if exploited: Interface hijack, user impersonation. Detection guidance: DOM event monitor, web logs. Hardening guidance: Sanitize stored captions. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"xss",
"iot / embedded devices",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Assess this network session and recommend defensive next steps. | session_id=SID_00965, protocol=TCP, packet_size=482, login_attempts=5, failed_logins=2, session_duration=1455.7s, encryption=None, ip_reputation_score=0.44, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=482, login_attempts=5, failed_logins=2, unusual_time_access=no, encryption=None, reputation_score=0.44, browser=Chrome, session_duration=1455.7s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Assess this network session and recommend defensive next steps. | session_id=SID_06889, protocol=TCP, packet_size=373, login_attempts=1, failed_logins=0, session_duration=317.4s, encryption=DES, ip_reputation_score=0.52, browser=Firefox, unusual_time_access=yes | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=373, login_attempts=1, failed_logins=0, unusual_time_access=yes, encryption=DES, reputation_score=0.52, browser=Firefox, session_duration=317.4s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=High, protocol=UDP, traffic_type=HTTP, packet_length=951, anomaly_score=61.98, attack_signature=Known Pattern B, malware_indicators=, alerts=Alert Triggered, action_taken=Logged, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with high-priority incident candidate. Evidence: severity=High, protocol=UDP, traffic_type=HTTP, packet_length=951, anomaly_score=61.98, signature=Known Pattern B, indicators=, alerts=Alert Triggered, current_action=Logged, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Open Fake AP, category=Network Security, attack_type=Open Wi-Fi Phishing / Data Intercept, scenario=Sets up a completely open AP with a common or generic SSID to attract users. Once connected, attacker can inspect traffic or launch phishing attacks., target_type=Any Wi-Fi Users, vulnerability=Open network, no auth, auto-connect, mitre_technique=T1185 – Man-in-the-Middle via Open Wi-Fi | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Open Fake AP is an Open Wi-Fi Phishing / Data Intercept scenario in Network Security affecting Any Wi-Fi Users. Primary weakness: Open network, no auth, auto-connect. MITRE mapping: T1185 – Man-in-the-Middle via Open Wi-Fi. Expected impact if exploited: Phishing, credential capture, traffic snooping. Detection guidance: Detect unauthorized open SSIDs; log rogue AP MACs; notify on new unknown APs. Hardening guidance: Avoid connecting to unknown open networks; use mobile data/VPN in public; disable auto-connect to open Wi-Fi. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"open wi-fi phishing / data in...",
"network security",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=To: [redacted-email] ############################################ # # # Adult Online Super Store # # Shhhhhh... # # You just found the Internet's # # BEST KEPT SECRET!!! # # 3 Vivid DVDs Absolutely FREE!!! # # # ############################################If you haven't seen it yet... we are offering a new weekly selection absolutely FREE Adult DVDs AND VIDEOS NO PURCHASE NECESSARY!!!!! GO TO: [redacted-url] must... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Hidden VBS Script in Startup, category=Insider Threat, attack_type=File Drop, scenario=Insider creates a .vbs file that runs silently and places it in the startup folder., target_type=Windows, vulnerability=Lack of script auditing in Startup, mitre_technique=T1059.005 (VBS Execution) | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Hidden VBS Script in Startup is an File Drop scenario in Insider Threat affecting Windows. Primary weakness: Lack of script auditing in Startup. MITRE mapping: T1059.005 (VBS Execution). Expected impact if exploited: Background info exfiltration. Detection guidance: Script path monitoring. Hardening guidance: Disable script engines, restrict folder access. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"file drop",
"insider threat",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=DDoS, severity=High, protocol=ICMP, traffic_type=DNS, packet_length=1000, anomaly_score=42.65, attack_signature=Known Pattern B, malware_indicators=, alerts=, action_taken=Blocked, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: DDoS telemetry with high-priority incident candidate. Evidence: severity=High, protocol=ICMP, traffic_type=DNS, packet_length=1000, anomaly_score=42.65, signature=Known Pattern B, indicators=, alerts=, current_action=Blocked, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"ddos",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Assess this network session and recommend defensive next steps. | session_id=SID_00932, protocol=UDP, packet_size=876, login_attempts=6, failed_logins=2, session_duration=2314.3s, encryption=DES, ip_reputation_score=0.27, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=UDP, packet_size=876, login_attempts=6, failed_logins=2, unusual_time_access=no, encryption=DES, reputation_score=0.27, browser=Chrome, session_duration=2314.3s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"udp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=secret page of maniac ! ! exalting ornament ! sapphire , notarize working staunchest . blight ! carlton , allison reformer lockers . i had no idea what that stupid chick was doing in the dark park . girls like her usually sit at home and read smart books . but she was brave enough to head for the farthest exit alone in the darkness . and she was dumb enough to resist me when i grabbed her . i spent too much time c... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=DDoS, severity=Low, protocol=ICMP, traffic_type=DNS, packet_length=770, anomaly_score=64.83, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=Alert Triggered, action_taken=Logged, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: DDoS telemetry with elevated event that merits analyst review. Evidence: severity=Low, protocol=ICMP, traffic_type=DNS, packet_length=770, anomaly_score=64.83, signature=Known Pattern B, indicators=IoC Detected, alerts=Alert Triggered, current_action=Logged, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"ddos",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Assess this network session and recommend defensive next steps. | session_id=SID_08699, protocol=UDP, packet_size=479, login_attempts=6, failed_logins=2, session_duration=132.8s, encryption=DES, ip_reputation_score=0.36, browser=Firefox, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=UDP, packet_size=479, login_attempts=6, failed_logins=2, unusual_time_access=no, encryption=DES, reputation_score=0.36, browser=Firefox, session_duration=132.8s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"udp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Assess this network session and recommend defensive next steps. | session_id=SID_07019, protocol=UDP, packet_size=758, login_attempts=3, failed_logins=1, session_duration=713.7s, encryption=None, ip_reputation_score=0.15, browser=Edge, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=UDP, packet_size=758, login_attempts=3, failed_logins=1, unusual_time_access=no, encryption=None, reputation_score=0.15, browser=Edge, session_duration=713.7s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"udp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=best defense against spam if you receive a vast amount of bulk email then please read on . this email was sent to you so you may stop unwanted emails from entering your inbox . let me tell you a little bit about myself and why i am imposing on your valuable time . my name is rich and i have been selling products on the internet for years . can you imagine the amount of spam i used to receive daily after years of d... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Assess this network session and recommend defensive next steps. | session_id=SID_06972, protocol=UDP, packet_size=489, login_attempts=5, failed_logins=1, session_duration=2269.6s, encryption=DES, ip_reputation_score=0.21, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: low. Evidence: protocol=UDP, packet_size=489, login_attempts=5, failed_logins=1, unusual_time_access=no, encryption=DES, reputation_score=0.21, browser=Chrome, session_duration=2269.6s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"udp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Assess this network session and recommend defensive next steps. | session_id=SID_07706, protocol=TCP, packet_size=440, login_attempts=4, failed_logins=2, session_duration=1696.2s, encryption=DES, ip_reputation_score=0.31, browser=Edge, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=TCP, packet_size=440, login_attempts=4, failed_logins=2, unusual_time_access=no, encryption=DES, reputation_score=0.31, browser=Edge, session_duration=1696.2s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=DDoS, severity=Medium, protocol=TCP, traffic_type=DNS, packet_length=1490, anomaly_score=81.62, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=, action_taken=Logged, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: DDoS telemetry with high-priority incident candidate. Evidence: severity=Medium, protocol=TCP, traffic_type=DNS, packet_length=1490, anomaly_score=81.62, signature=Known Pattern B, indicators=IoC Detected, alerts=, current_action=Logged, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"ddos",
"tcp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=Medium, protocol=ICMP, traffic_type=DNS, packet_length=898, anomaly_score=95.51, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=Alert Triggered, action_taken=Ignored, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with high-priority incident candidate. Evidence: severity=Medium, protocol=ICMP, traffic_type=DNS, packet_length=898, anomaly_score=95.51, signature=Known Pattern B, indicators=IoC Detected, alerts=Alert Triggered, current_action=Ignored, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Mshta Execution with Remote HTA, category=Red Team, attack_type=HTA Execution via mshta.exe, scenario=Executes malicious script using mshta.exe, which can run remote .hta or embedded VBScript., target_type=Endpoints, vulnerability=Execution from remote URL, mitre_technique=T1218.005 – Mshta | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Mshta Execution with Remote HTA is an HTA Execution via mshta.exe scenario in Red Team affecting Endpoints. Primary weakness: Execution from remote URL. MITRE mapping: T1218.005 – Mshta. Expected impact if exploited: Stealthy in-memory execution of scripts. Detection guidance: Monitor mshta with command-line logging. Hardening guidance: Block mshta in enterprise via GPO. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"hta execution via mshta.exe",
"red team",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Assess this network session and recommend defensive next steps. | session_id=SID_04942, protocol=TCP, packet_size=458, login_attempts=5, failed_logins=2, session_duration=1021.1s, encryption=AES, ip_reputation_score=0.32, browser=Edge, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=458, login_attempts=5, failed_logins=2, unusual_time_access=no, encryption=AES, reputation_score=0.32, browser=Edge, session_duration=1021.1s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=High, protocol=ICMP, traffic_type=FTP, packet_length=835, anomaly_score=43.18, attack_signature=Known Pattern B, malware_indicators=, alerts=, action_taken=Logged, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with high-priority incident candidate. Evidence: severity=High, protocol=ICMP, traffic_type=FTP, packet_length=835, anomaly_score=43.18, signature=Known Pattern B, indicators=, alerts=, current_action=Logged, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=failure notice hi . this is the qmail - send program at www . servernl . net . i ' m afraid i wasn ' t able to deliver your message to the following addresses . this is a permanent error ; i ' ve given up . sorry it didn ' t work out . : this address no longer accepts mail . - - - below this line is a copy of the message . return - path : received : ( qmail 6891 invoked from network ) ; 19 jul 2005 11 : 16 : 06 -... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=DDoS, severity=Low, protocol=UDP, traffic_type=HTTP, packet_length=963, anomaly_score=29.34, attack_signature=Known Pattern B, malware_indicators=, alerts=Alert Triggered, action_taken=Ignored, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: DDoS telemetry with lower-severity signal that still needs validation. Evidence: severity=Low, protocol=UDP, traffic_type=HTTP, packet_length=963, anomaly_score=29.34, signature=Known Pattern B, indicators=, alerts=Alert Triggered, current_action=Ignored, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"ddos",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=get your prescript % ion delivered to your door ! e psychogenetics unimbued dome overspeculate gustation illicitly docking misbelieve protoconulid scuta . botuliform impartial isonephelic wheelhorse legitim saron impawned bulletproof glycerophosphoric . franchisers pullers sparged grandchild ignatia . isonephelic oysterhood broadaxe curiologically pyritology jacket maar weaponries hijacked . interpolater aspartyl... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=DDoS, severity=High, protocol=UDP, traffic_type=HTTP, packet_length=1459, anomaly_score=87.81, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=Alert Triggered, action_taken=Ignored, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: DDoS telemetry with high-priority incident candidate. Evidence: severity=High, protocol=UDP, traffic_type=HTTP, packet_length=1459, anomaly_score=87.81, signature=Known Pattern B, indicators=IoC Detected, alerts=Alert Triggered, current_action=Ignored, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"ddos",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Assess this network session and recommend defensive next steps. | session_id=SID_01875, protocol=TCP, packet_size=762, login_attempts=6, failed_logins=0, session_duration=223.6s, encryption=None, ip_reputation_score=0.25, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: low. Evidence: protocol=TCP, packet_size=762, login_attempts=6, failed_logins=0, unusual_time_access=no, encryption=None, reputation_score=0.25, browser=Chrome, session_duration=223.6s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=Medium, protocol=UDP, traffic_type=FTP, packet_length=572, anomaly_score=44.74, attack_signature=Known Pattern B, malware_indicators=, alerts=, action_taken=Ignored, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with elevated event that merits analyst review. Evidence: severity=Medium, protocol=UDP, traffic_type=FTP, packet_length=572, anomaly_score=44.74, signature=Known Pattern B, indicators=, alerts=, current_action=Ignored, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=dont pay more than $ 100 for ur softwares miseries chheap softtwares for you , all are original oemmajor titles from micros 0 ft and adobe for rock bottom prriicegreat bargaain sale ! variety discoount softtwares at wholesale chaeap pricing ! microsoft windows xp professional - my priice : $ 50 normal priice : $ 270 . 99 ; you savve $ 220 microsoft office xp professional - my priice : $ 100 normal priice : $ 579 .... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Malware, severity=Medium, protocol=UDP, traffic_type=HTTP, packet_length=884, anomaly_score=70.91, attack_signature=Known Pattern A, malware_indicators=, alerts=Alert Triggered, action_taken=Blocked, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Malware telemetry with elevated event that merits analyst review. Evidence: severity=Medium, protocol=UDP, traffic_type=HTTP, packet_length=884, anomaly_score=70.91, signature=Known Pattern A, indicators=, alerts=Alert Triggered, current_action=Blocked, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"malware",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Turn this cyber incident trend into a defensive planning brief. | country=China, year=2023, attack_type=Man-in-the-Middle, target_industry=Education, vulnerability=Zero-day, defense_used=Antivirus, attack_source=Hacker Group, financial_loss_musd=35.37, affected_users=455900, resolution_hours=11 | Provide risk prioritization, hardening guidance, and monitoring recommendations for a blue-team audience. | Risk summary: Man-in-the-Middle against the Education sector in China shows a high defensive planning priority. Observed weaknesses center on Zero-day, with reported source attribution of Hacker Group; existing control listed was Antivirus. Planning actions: 1) review whether your environment has the same exposure class and close the weakness through patching, stronger identity controls, or phishing resistance as appropriate, 2) ensure detections cover early indicators for man-in-the-middle activity and escalation paths in the education workflow, 3) prepare business-impact communications because reported impact reached $35.37M and 455900 affected users, 4) shorten containment and recovery playbooks if current resolution times approach 11 hours. | [
"threat-intel",
"risk-planning",
"man-in-the-middle",
"education"
] | atharvasoundankar/global-cybersecurity-threats-2015-2024 | 0.72 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Wi-Fi Flood Near QKD Control Channel, category=Quantum Cryptography & Post-Quantum Threats, attack_type=Wireless Protocol Flooding, scenario=Overloads classical communication control channel of QKD with 802.11 flooding, preventing coordination between endpoints., target_type=QKD over Hybrid Wireless, vulnerability=Wi-Fi Control Channel Congestion, mitre_technique=T0814 - Wireless Protocol Manipulation | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Wi-Fi Flood Near QKD Control Channel is an Wireless Protocol Flooding scenario in Quantum Cryptography & Post-Quantum Threats affecting QKD over Hybrid Wireless. Primary weakness: Wi-Fi Control Channel Congestion. MITRE mapping: T0814 - Wireless Protocol Manipulation. Expected impact if exploited: Disruption of classical coordination. Detection guidance: Wi-Fi traffic anomaly detection; Packet loss monitoring. Hardening guidance: Use wired out-of-band control channels or hardened wireless. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"wireless protocol flooding",
"quantum cryptography & post-q...",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=Hi There,I recently came across a copy of your resume I found on the internet. In case you're back on the market, you should check out this new web service that instantly posts your resume to over 75 career websites.You fill out one online form and immediately you can be seen by over 1.5 million employers and recruiters daily! It works.In about 15 minutes you'll be posted to all the major sites like: Monster, Care... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Assess this network session and recommend defensive next steps. | session_id=SID_04292, protocol=ICMP, packet_size=413, login_attempts=6, failed_logins=2, session_duration=153.5s, encryption=AES, ip_reputation_score=0.09, browser=Chrome, unusual_time_access=yes | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=ICMP, packet_size=413, login_attempts=6, failed_logins=2, unusual_time_access=yes, encryption=AES, reputation_score=0.09, browser=Chrome, session_duration=153.5s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) verify whether the access time aligns with approved maintenance or travel activity. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"icmp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Assess this network session and recommend defensive next steps. | session_id=SID_09053, protocol=TCP, packet_size=485, login_attempts=3, failed_logins=1, session_duration=356.5s, encryption=AES, ip_reputation_score=0.32, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: low. Evidence: protocol=TCP, packet_size=485, login_attempts=3, failed_logins=1, unusual_time_access=no, encryption=AES, reputation_score=0.32, browser=Chrome, session_duration=356.5s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Assess this network session and recommend defensive next steps. | session_id=SID_03442, protocol=UDP, packet_size=377, login_attempts=4, failed_logins=1, session_duration=1759.1s, encryption=DES, ip_reputation_score=0.48, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: low. Evidence: protocol=UDP, packet_size=377, login_attempts=4, failed_logins=1, unusual_time_access=no, encryption=DES, reputation_score=0.48, browser=Chrome, session_duration=1759.1s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"udp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=confidential business confidential business i presume this mail will not be a surprise to you . i am an accountant with the ministry of mineral resources and energy in south africa and also a member of contracts awarding committee of this ministry under south africa government . many years ago , south africa government asked this committee to awards contracts to foreign firms , which i and two of my partners are t... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=Low, protocol=UDP, traffic_type=DNS, packet_length=834, anomaly_score=7.49, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=Alert Triggered, action_taken=Blocked, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with lower-severity signal that still needs validation. Evidence: severity=Low, protocol=UDP, traffic_type=DNS, packet_length=834, anomaly_score=7.49, signature=Known Pattern B, indicators=IoC Detected, alerts=Alert Triggered, current_action=Blocked, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Network Fuzzing of Custom IoT Protocol Using AFLnet, category=Zero-Day Research / Fuzzing, attack_type=Fuzzer Configuration, scenario=Fuzzing an IoT device’s proprietary TCP protocol using AFLnet., target_type=Network Service, vulnerability=Protocol fuzzing, mitre_technique=T1046 | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Network Fuzzing of Custom IoT Protocol Using AFLnet is an Fuzzer Configuration scenario in Zero-Day Research / Fuzzing affecting Network Service. Primary weakness: Protocol fuzzing. MITRE mapping: T1046. Expected impact if exploited: Device DoS or takeover. Detection guidance: Device logs, AFLnet crash data. Hardening guidance: Harden protocol parsing. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"fuzzer configuration",
"zero-day research / fuzzing",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=> - Joe > > P.S. I hate everybody.one up: I hate everything post-(the first ten seconds of the universe), but especially everybody and everything, and including but not limited to puppies and ice cream. [redacted-url] | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Quantum Channel Mapping via Wireless Signal Reflection, category=Quantum Cryptography & Post-Quantum Threats, attack_type=Quantum Trojan Horse via Spatial Reflection, scenario=The attacker uses radar-like methods to map physical QKD fiber layouts and vulnerabilities., target_type=QKD Infrastructure, vulnerability=Exposed fiber layout; no channel obfuscation, mitre_technique=T1595.002 (Active Scanning) | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Quantum Channel Mapping via Wireless Signal Reflection is an Quantum Trojan Horse via Spatial Reflection scenario in Quantum Cryptography & Post-Quantum Threats affecting QKD Infrastructure. Primary weakness: Exposed fiber layout; no channel obfuscation. MITRE mapping: T1595.002 (Active Scanning). Expected impact if exploited: Reconnaissance aiding future physical compromise. Detection guidance: Radar signature analysis; EM field monitoring. Hardening guidance: Conceal fiber layout; randomized pathing and shielding. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"quantum trojan horse via spat...",
"quantum cryptography & post-q...",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Malware, severity=Medium, protocol=UDP, traffic_type=HTTP, packet_length=1377, anomaly_score=72.70, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=, action_taken=Logged, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Malware telemetry with elevated event that merits analyst review. Evidence: severity=Medium, protocol=UDP, traffic_type=HTTP, packet_length=1377, anomaly_score=72.70, signature=Known Pattern B, indicators=IoC Detected, alerts=, current_action=Logged, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"malware",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=reminder ~ alexis de tocqueville breakfast reminder ! the alexis de tocqueville breakfast is just a few days away . please rsvp if you have not already done so . date : tuesday , july 31 time : 7 : 30 - 9 : 00 a . m . location : the center for the retarded , 3550 w . dallas , between waugh dr . and shepherd rsvp : please call or e - mail zulie flores , ext . 3 - 3908 transportation : a bus will depart from the and... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=High, protocol=ICMP, traffic_type=FTP, packet_length=911, anomaly_score=27.60, attack_signature=Known Pattern A, malware_indicators=IoC Detected, alerts=Alert Triggered, action_taken=Logged, log_source=Firewall | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with high-priority incident candidate. Evidence: severity=High, protocol=ICMP, traffic_type=FTP, packet_length=911, anomaly_score=27.60, signature=Known Pattern A, indicators=IoC Detected, alerts=Alert Triggered, current_action=Logged, source_system=Firewall. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=she is your wife in the video goodbye billion meccabask galt covenantelectra gangplank netherlandscolby regretful odiumdispensable zoroastrian papuawalcott inaccuracy pricecoagulate polytypy joustdid emphases astatineaudacity berniece dichotomyconjoint | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Assess this network session and recommend defensive next steps. | session_id=SID_04890, protocol=TCP, packet_size=627, login_attempts=1, failed_logins=0, session_duration=1253.0s, encryption=None, ip_reputation_score=0.16, browser=Edge, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=627, login_attempts=1, failed_logins=0, unusual_time_access=no, encryption=None, reputation_score=0.16, browser=Edge, session_duration=1253.0s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Assess this network session and recommend defensive next steps. | session_id=SID_07528, protocol=TCP, packet_size=419, login_attempts=6, failed_logins=3, session_duration=127.5s, encryption=DES, ip_reputation_score=0.16, browser=Firefox, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=TCP, packet_size=419, login_attempts=6, failed_logins=3, unusual_time_access=no, encryption=DES, reputation_score=0.16, browser=Firefox, session_duration=127.5s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Turn this cyber incident trend into a defensive planning brief. | country=Japan, year=2016, attack_type=Man-in-the-Middle, target_industry=IT, vulnerability=Unpatched Software, defense_used=Antivirus, attack_source=Insider, financial_loss_musd=57.14, affected_users=590262, resolution_hours=15 | Provide risk prioritization, hardening guidance, and monitoring recommendations for a blue-team audience. | Risk summary: Man-in-the-Middle against the IT sector in Japan shows a high defensive planning priority. Observed weaknesses center on Unpatched Software, with reported source attribution of Insider; existing control listed was Antivirus. Planning actions: 1) review whether your environment has the same exposure class and close the weakness through patching, stronger identity controls, or phishing resistance as appropriate, 2) ensure detections cover early indicators for man-in-the-middle activity and escalation paths in the it workflow, 3) prepare business-impact communications because reported impact reached $57.14M and 590262 affected users, 4) shorten containment and recovery playbooks if current resolution times approach 15 hours. | [
"threat-intel",
"risk-planning",
"man-in-the-middle",
"it"
] | atharvasoundankar/global-cybersecurity-threats-2015-2024 | 0.72 |
Assess this network session and recommend defensive next steps. | session_id=SID_05280, protocol=TCP, packet_size=515, login_attempts=4, failed_logins=0, session_duration=1432.5s, encryption=AES, ip_reputation_score=0.33, browser=Chrome, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: low. Evidence: protocol=TCP, packet_size=515, login_attempts=4, failed_logins=0, unusual_time_access=no, encryption=AES, reputation_score=0.33, browser=Chrome, session_duration=1432.5s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Malware, severity=High, protocol=TCP, traffic_type=DNS, packet_length=547, anomaly_score=72.61, attack_signature=Known Pattern B, malware_indicators=, alerts=Alert Triggered, action_taken=Logged, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Malware telemetry with high-priority incident candidate. Evidence: severity=High, protocol=TCP, traffic_type=DNS, packet_length=547, anomaly_score=72.61, signature=Known Pattern B, indicators=, alerts=Alert Triggered, current_action=Logged, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"malware",
"tcp"
] | teamincribo/cyber-security-attacks | 0.76 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=DDoS, severity=Low, protocol=UDP, traffic_type=DNS, packet_length=662, anomaly_score=21.88, attack_signature=Known Pattern B, malware_indicators=IoC Detected, alerts=, action_taken=Logged, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: DDoS telemetry with lower-severity signal that still needs validation. Evidence: severity=Low, protocol=UDP, traffic_type=DNS, packet_length=662, anomaly_score=21.88, signature=Known Pattern B, indicators=IoC Detected, alerts=, current_action=Logged, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"ddos",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Cronjob for Data Exfiltration Script, category=Red Team, attack_type=Scheduled Exfiltration Job, scenario=Automates periodic sending of logs or stolen data., target_type=Linux Servers, vulnerability=Cron allows silent data movement, mitre_technique=T1030 – Data Transfer Size Limits | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Cronjob for Data Exfiltration Script is an Scheduled Exfiltration Job scenario in Red Team affecting Linux Servers. Primary weakness: Cron allows silent data movement. MITRE mapping: T1030 – Data Transfer Size Limits. Expected impact if exploited: Periodic data leakage to external server. Detection guidance: Monitor network traffic on schedule. Hardening guidance: Restrict outbound cron scripts. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"scheduled exfiltration job",
"red team",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Assess this network session and recommend defensive next steps. | session_id=SID_07252, protocol=TCP, packet_size=352, login_attempts=4, failed_logins=1, session_duration=166.8s, encryption=AES, ip_reputation_score=0.19, browser=Firefox, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=352, login_attempts=4, failed_logins=1, unusual_time_access=no, encryption=AES, reputation_score=0.19, browser=Firefox, session_duration=166.8s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Intrusion, severity=High, protocol=ICMP, traffic_type=HTTP, packet_length=1292, anomaly_score=85.35, attack_signature=Known Pattern A, malware_indicators=, alerts=, action_taken=Ignored, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Intrusion telemetry with high-priority incident candidate. Evidence: severity=High, protocol=ICMP, traffic_type=HTTP, packet_length=1292, anomaly_score=85.35, signature=Known Pattern A, indicators=, alerts=, current_action=Ignored, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"intrusion",
"icmp"
] | teamincribo/cyber-security-attacks | 0.76 |
Assess this network session and recommend defensive next steps. | session_id=SID_05112, protocol=TCP, packet_size=1007, login_attempts=5, failed_logins=0, session_duration=2766.4s, encryption=DES, ip_reputation_score=0.33, browser=Safari, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: likely malicious session requiring triage. Risk level: high. Evidence: protocol=TCP, packet_size=1007, login_attempts=5, failed_logins=0, unusual_time_access=no, encryption=DES, reputation_score=0.33, browser=Safari, session_duration=2766.4s. Recommended actions: 1) apply temporary containment such as step-up auth, source blocking, or host isolation if business impact allows, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.8 |
Assess this network session and recommend defensive next steps. | session_id=SID_02547, protocol=TCP, packet_size=224, login_attempts=3, failed_logins=2, session_duration=89.4s, encryption=None, ip_reputation_score=0.54, browser=Unknown, unusual_time_access=yes | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=224, login_attempts=3, failed_logins=2, unusual_time_access=yes, encryption=None, reputation_score=0.54, browser=Unknown, session_duration=89.4s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Plugin Auto-Update Attack, category=Mobile Security, attack_type=Plugin author is compromised → auto-updates agent tools, scenario=LangChain plugin auto-update logic causes previously safe plugin to silently update to a malicious version after author’s account is compromised., target_type=Agent Plugin Auto-Update Logic, vulnerability=Auto-update from compromised author source, mitre_technique=T1195.002 – Supply Chain Compromise | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Plugin Auto-Update Attack is an Plugin author is compromised → auto-updates agent tools scenario in Mobile Security affecting Agent Plugin Auto-Update Logic. Primary weakness: Auto-update from compromised author source. MITRE mapping: T1195.002 – Supply Chain Compromise. Expected impact if exploited: Silent agent hijack, prompt/token leak on update. Detection guidance: Monitor update frequency, changes in plugin behavior; log plugin hashes pre- and post-update. Hardening guidance: Lock plugin versions; disable auto-updates; require manual review for all plugin upgrades. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"plugin author is compromised...",
"mobile security",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=germanic linguistics & philology the web site of the society for germanic philology ( sgp ) has moved to the following address : www . germanic . ohio-state . edu / sgp / the sgp embraces all areas of and approaches to germanic linguistics and philology , from formal syntax and phonology through historical linguistics to textual editing and includes scholars interested in all germanic languages , from modern germa... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely benign or routine correspondence. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: still verify sender authenticity, but keep the message in monitoring-only posture unless header, domain, or attachment checks reveal additional risk. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.68 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Lack of API Response Throttling Enables Brute Force Device Discovery, category=IoT / Embedded Devices, attack_type=Device Enumeration, scenario=An attacker brute-forces thousands of device IDs using API requests to discover active smart devices., target_type=Smart Sensors / Meters, vulnerability=Unthrottled and predictable API queries, mitre_technique=T1595 (Active Scanning) | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Lack of API Response Throttling Enables Brute Force Device Discovery is an Device Enumeration scenario in IoT / Embedded Devices affecting Smart Sensors / Meters. Primary weakness: Unthrottled and predictable API queries. MITRE mapping: T1595 (Active Scanning). Expected impact if exploited: Privacy breach, targeted attacks. Detection guidance: Rate-limit failed or repeated API requests. Hardening guidance: Introduce CAPTCHAs, API tokens, and rate control. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"device enumeration",
"iot / embedded devices",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this email from a security operations perspective and give the next response steps. | email_excerpt=has your cum ever dribbled and you wish it had shot out ? heya ! has your cum ever dribbled and you wish it had shot out ? have you ever wanted to impress your girl with a huge cumshot ? spur - m is the only site to offer an all natural male enhancement formula that is proven to increase your sperm volume by up to 500 % . our highly potent , volume enhancing formula will give our results in days and comes with an... | Use a defensive email triage lens. Focus on phishing indicators, user safety, and tenant-wide follow-up. | Classification: likely phishing. Analyst reasoning: review the sender claims, urgency cues, request type, links or attachment behavior, and whether the message content matches normal business processes. Recommended response: quarantine the message, preserve headers, search for matching sender and domains across the tenant, block indicators if confirmed, and reset credentials if the recipient interacted. User guidance: do not click links or open attachments until the message is cleared, and document the verdict plus notable indicators for future mail filtering. | [
"email-security",
"phishing-triage",
"classification",
"user-awareness"
] | subhajournal/phishingemails | 0.74 |
Turn this cyber incident trend into a defensive planning brief. | country=Japan, year=2020, attack_type=Man-in-the-Middle, target_industry=Education, vulnerability=Unpatched Software, defense_used=AI-based Detection, attack_source=Insider, financial_loss_musd=29.58, affected_users=141411, resolution_hours=58 | Provide risk prioritization, hardening guidance, and monitoring recommendations for a blue-team audience. | Risk summary: Man-in-the-Middle against the Education sector in Japan shows a moderate defensive planning priority. Observed weaknesses center on Unpatched Software, with reported source attribution of Insider; existing control listed was AI-based Detection. Planning actions: 1) review whether your environment has the same exposure class and close the weakness through patching, stronger identity controls, or phishing resistance as appropriate, 2) ensure detections cover early indicators for man-in-the-middle activity and escalation paths in the education workflow, 3) prepare business-impact communications because reported impact reached $29.58M and 141411 affected users, 4) shorten containment and recovery playbooks if current resolution times approach 58 hours. | [
"threat-intel",
"risk-planning",
"man-in-the-middle",
"education"
] | atharvasoundankar/global-cybersecurity-threats-2015-2024 | 0.72 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Malware, severity=High, protocol=UDP, traffic_type=FTP, packet_length=635, anomaly_score=17.44, attack_signature=Known Pattern B, malware_indicators=, alerts=Alert Triggered, action_taken=Blocked, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Malware telemetry with high-priority incident candidate. Evidence: severity=High, protocol=UDP, traffic_type=FTP, packet_length=635, anomaly_score=17.44, signature=Known Pattern B, indicators=, alerts=Alert Triggered, current_action=Blocked, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"malware",
"udp"
] | teamincribo/cyber-security-attacks | 0.76 |
Explain this application security scenario and provide defensive detection and hardening guidance. | title=Remote Desktop Lockout, category=Insider Threat, attack_type=User Lockout Sabotage, scenario=Employee logs into remote desktop systems and locks them continuously, preventing others from working., target_type=Remote Systems, vulnerability=No session timeout policies, mitre_technique=T1531 (Account Access Removal) | Stay on defensive guidance only. Focus on prevention, monitoring, validation, and remediation. | Defensive summary: Remote Desktop Lockout is an User Lockout Sabotage scenario in Insider Threat affecting Remote Systems. Primary weakness: No session timeout policies. MITRE mapping: T1531 (Account Access Removal). Expected impact if exploited: Work interruption. Detection guidance: Session log analysis. Hardening guidance: Use RDP timeout and force logout. Operational next step: verify exposed entry points, add regression tests for the vulnerable pattern, and confirm the mitigation is enforced in CI or deployment gates. | [
"appsec",
"user lockout sabotage",
"insider threat",
"defensive-guidance"
] | tannubarot/cybersecurity-attack-and-defence-dataset | 0.77 |
Classify this security telemetry and provide immediate SOC response actions. | attack_type=Malware, severity=High, protocol=TCP, traffic_type=HTTP, packet_length=605, anomaly_score=5.47, attack_signature=Known Pattern A, malware_indicators=IoC Detected, alerts=, action_taken=Blocked, log_source=Server | Treat the record as a defensive triage prompt. Do not provide offensive reproduction steps. | Classification: Malware telemetry with high-priority incident candidate. Evidence: severity=High, protocol=TCP, traffic_type=HTTP, packet_length=605, anomaly_score=5.47, signature=Known Pattern A, indicators=IoC Detected, alerts=, current_action=Blocked, source_system=Server. Immediate response: 1) confirm whether the current action was effective and escalate if blocked/ignored controls were insufficient, 2) correlate firewall, IDS/IPS, endpoint, and authentication logs for related activity, 3) scope impacted hosts or segments and apply containment where confidence justifies it, 4) tune detections around the observed signature and anomaly profile, 5) record the disposition and lessons learned for future triage. | [
"soc",
"telemetry-triage",
"malware",
"tcp"
] | teamincribo/cyber-security-attacks | 0.76 |
Assess this network session and recommend defensive next steps. | session_id=SID_05051, protocol=TCP, packet_size=645, login_attempts=2, failed_logins=1, session_duration=1576.8s, encryption=None, ip_reputation_score=0.16, browser=Safari, unusual_time_access=no | Classify risk conservatively, stay defensive, and focus on containment, validation, and logging improvements. | Assessment: mixed signal session that needs validation before escalation. Risk level: medium. Evidence: protocol=TCP, packet_size=645, login_attempts=2, failed_logins=1, unusual_time_access=no, encryption=None, reputation_score=0.16, browser=Safari, session_duration=1576.8s. Recommended actions: 1) keep the session under heightened monitoring while confirming owner, device, and change window context, 2) correlate this session with authentication, proxy, and endpoint telemetry, 3) review neighboring sessions from the same principal or host for repetition and spread, 4) treat weak or absent encryption as a hardening gap and enforce stronger transport controls. | [
"intrusion-detection",
"triage",
"defensive-netsec",
"tcp"
] | dnkumars/cybersecurity-intrusion-detection-dataset | 0.7 |
End of preview.
README.md exists but content is empty.
- Downloads last month
- 2