Upload folder using huggingface_hub

Dockerfile

@@ -0,0 +1,17 @@
+# Use an official Python runtime as a parent image
+FROM python:3.9-slim
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Copy the current directory contents into the container at /app
+COPY . .
+
+# Install any needed packages specified in requirements.txt
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Make port 7860 available to the world outside this container
+EXPOSE 7860
+
+# Run app.py when the container launches
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,10 +1,12 @@
----
-title: HFGitDorker
-emoji: 🏢
-colorFrom: gray
-colorTo: gray
-sdk: docker
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---
+title: HF GitDorker
+emoji: 🕵️
+colorFrom: blue
+colorTo: green
+sdk: docker
+app_port: 7860
+---
+
+# HF GitDorker
+
+A custom-built, real-time GitHub dorking tool designed for Hugging Face Spaces.

app.py

@@ -0,0 +1,42 @@
+import os
+import asyncio
+from fastapi import FastAPI, Query, Request
+from fastapi.responses import HTMLResponse, StreamingResponse
+from fastapi.templating import Jinja2Templates
+from dorker import search
+
+app = FastAPI()
+templates = Jinja2Templates(directory="templates")
+
+async def stream_dorker_results(tokens, query, dork_file_path):
+    """Wraps the dorker.search generator and formats output for SSE."""
+    try:
+        with open(dork_file_path, 'r', encoding='utf-8') as f:
+            dorks = [line.strip() for line in f if line.strip()]
+    except FileNotFoundError:
+        yield f"data: [ERROR] Dorks file not found: {dork_file_path}\n\n"
+        yield f"data: [STREAM_COMPLETE]\n\n"
+        return
+
+    async for result_line in search(tokens, query, dorks):
+        yield f"data: {result_line}\n\n"
+        await asyncio.sleep(0.01)
+
+    yield f"data: [STREAM_COMPLETE]\n\n"
+
+@app.get("/", response_class=HTMLResponse)
+async def read_root(request: Request):
+    return templates.TemplateResponse("index.html", {"request": request})
+
+@app.get("/run", response_class=StreamingResponse)
+async def run_dorker_stream(query: str = Query(...), dork_file: str = Query(...)):
+    tokens_str = os.environ.get("GHA_TOKENS", "")
+    tokens = [token.strip() for token in tokens_str.split(',') if token.strip()]
+
+    if not tokens:
+        async def error_generator():
+            yield "data: [ERROR] GHA_TOKENS is not set in the Space secrets.\n\n"
+            yield "data: [STREAM_COMPLETE]\n\n"
+        return StreamingResponse(error_generator(), media_type="text/event-stream")
+
+    return StreamingResponse(stream_dorker_results(tokens, query, dork_file), media_type="text/event-stream")

dorker.py

@@ -0,0 +1,58 @@
+import asyncio
+import httpx
+from itertools import cycle
+
+GITHUB_API_URL = "https://api.github.com/search/code"
+
+def urlencode(s: str) -> str:
+    """A simple URL encoder for query strings."""
+    return s.replace(':', '%3A').replace('"', '%22').replace(' ', '+')
+
+async def search(tokens: list, query: str, dorks: list):
+    """
+    Async generator to search GitHub for dorks and stream results in real-time.
+    """
+    if not tokens:
+        yield "[ERROR] No GitHub tokens provided in GHA_TOKENS secret."
+        return
+
+    token_cycler = cycle(tokens)
+    headers = {"Accept": "application/vnd.github.v3+json"}
+    
+    async with httpx.AsyncClient() as client:
+        for i, dork in enumerate(dorks):
+            full_query = f"{query} {dork}"
+            url = f"{GITHUB_API_URL}?q={urlencode(full_query)}"
+            
+            current_token = next(token_cycler)
+            headers["Authorization"] = f"token {current_token}"
+
+            yield f"[INFO] [{i+1}/{len(dorks)}] Searching with dork: {dork}"
+            
+            try:
+                res = await client.get(url, headers=headers)
+
+                if res.status_code == 403 and 'rate limit' in res.text.lower():
+                    yield "[WARN] Rate limit hit. Sleeping for 60 seconds..."
+                    await asyncio.sleep(60)
+                    res = await client.get(url, headers=headers) # Retry
+
+                res.raise_for_status()
+                
+                data = res.json()
+                count = data.get("total_count", 0)
+                
+                github_search_url = f"https://github.com/search?q={urlencode(full_query)}&type=Code"
+
+                if count > 0:
+                    yield f"[+] FOUND ({count} results): {dork} -> {github_search_url}"
+                else:
+                    yield f"[-] Not Found: {dork}"
+
+            except httpx.HTTPStatusError as e:
+                yield f"[ERROR] HTTP Error for '{dork}': {e.response.status_code}"
+            except Exception as e:
+                yield f"[ERROR] Unexpected error for '{dork}': {str(e)}"
+            
+            # Sleep to stay within the 30 requests/minute limit
+            await asyncio.sleep(2.1)

dorks/alldorks.txt

@@ -0,0 +1,33 @@
+filename:.env
+filename:.bash_profile
+filename:.bashrc
+filename:.gitconfig
+filename:credentials
+filename:secrets
+filename:config.json
+
+NPM_TOKEN
+GITHUB_TOKEN
+HEROKU_API_KEY
+
+language:yaml "aws_access_key"
+
+"\.mlab.com" password
+
+extension:pem private
+
+[WFClient] Password=
+
+JEKYLL_GITHUB_TOKEN
+
+SF_USERNAME salesforce
+
+filename:sftp-config.json
+
+filename:idea14.key
+
+filename:hub oauth_token
+
+filename:dbeaver-data-sources.xml
+
+filename:LocalSettings.php

requirements.txt

@@ -0,0 +1,4 @@
+fastapi
+uvicorn
+jinja2
+httpx

templates/index.html

@@ -0,0 +1,94 @@
+<!DOCTYPE html>
+<html>
+<head>
+    <title>HF GitDorker</title>
+    <style>
+        body { font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, Helvetica, Arial, sans-serif; background-color: #f4f4f9; color: #333; margin: 2em; }
+        .container { max-width: 900px; margin: auto; background: white; padding: 2em; border-radius: 8px; box-shadow: 0 2px 4px rgba(0,0,0,0.1); }
+        h1 { color: #4a4a4a; }
+        form { display: flex; flex-direction: column; gap: 1em; margin-bottom: 2em;}
+        input[type=text], select { padding: 10px; border-radius: 4px; border: 1px solid #ccc; font-size: 1em; }
+        button { padding: 10px 20px; border: none; border-radius: 4px; background-color: #007bff; color: white; font-size: 1em; cursor: pointer; transition: background-color 0.2s; }
+        button:hover { background-color: #0056b3; }
+        button:disabled { background-color: #cccccc; cursor: not-allowed; }
+        #results-container h2 { margin-bottom: 0.5em; }
+        #results { min-height: 200px; max-height: 60vh; overflow-y: auto; background-color: #282c34; color: #abb2bf; padding: 1em; border-radius: 4px; white-space: pre-wrap; word-wrap: break-word; font-family: Menlo, Monaco, Consolas, "Courier New", monospace; font-size: 0.9em;}
+        #status { margin-top: 1em; font-style: italic; color: #555; }
+        .line-info { color: #61afef; }
+        .line-warn { color: #e5c07b; }
+        .line-error { color: #e06c75; }
+        .line-success { color: #98c379; }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>HF GitDorker</h1>
+        <form id="dorker-form">
+            <label for="query">Query (e.g., tesla.com):</label>
+            <input type="text" id="query" name="query" required>
+            
+            <label for="dork_file">Dorks File:</label>
+            <select id="dork_file" name="dork_file">
+                <option value="dorks/alldorks.txt">All Dorks</option>
+            </select>
+
+            <button type="submit" id="run-button">Run Dorker</button>
+        </form>
+        
+        <div id="results-container">
+            <h2>Live Output:</h2>
+            <pre id="results"></pre>
+            <p id="status"></p>
+        </div>
+    </div>
+
+    <script>
+        document.getElementById('dorker-form').addEventListener('submit', function(event) {
+            event.preventDefault();
+
+            const query = document.getElementById('query').value;
+            const dorkFile = document.getElementById('dork_file').value;
+            const resultsDiv = document.getElementById('results');
+            const statusDiv = document.getElementById('status');
+            const runButton = document.getElementById('run-button');
+
+            resultsDiv.innerHTML = '';
+            statusDiv.textContent = '🚀 Starting... please wait.';
+            runButton.disabled = true;
+
+            const evtSource = new EventSource(`/run?query=${encodeURIComponent(query)}&dork_file=${encodeURIComponent(dorkFile)}`);
+
+            evtSource.onmessage = function(event) {
+                const data = event.data;
+                if (data === '[STREAM_COMPLETE]') {
+                    evtSource.close();
+                    statusDiv.textContent = '✅ Done.';
+                    runButton.disabled = false;
+                    return;
+                }
+                
+                const line = document.createElement('span');
+                if (data.startsWith('[INFO]')) {
+                    line.className = 'line-info';
+                } else if (data.startsWith('[WARN]')) {
+                    line.className = 'line-warn';
+                } else if (data.startsWith('[ERROR]')) {
+                    line.className = 'line-error';
+                } else if (data.startsWith('[+]')) {
+                    line.className = 'line-success';
+                }
+                line.textContent = data + '\n';
+                resultsDiv.appendChild(line);
+                resultsDiv.scrollTop = resultsDiv.scrollHeight;
+            };
+
+            evtSource.onerror = function(err) {
+                console.error("EventSource failed:", err);
+                statusDiv.textContent = '❌ Error occurred. Connection closed. Check Space logs for details.';
+                runButton.disabled = false;
+                evtSource.close();
+            };
+        });
+    </script>
+</body>
+</html>