{"repo": "aradotso/security-skills", "files": {"skills/deepseek-pentest-ai-burp-extension/SKILL.md": {"1f6e1ee6e67149931170223ea6e94f72dbb821b7": {"t": 1785746535.2219322, "n": 12491, "h": "09509946e0143925", "c": "---\nname: deepseek-pentest-ai-burp-extension\ndescription: AI-powered Burp Suite extension for automated payload generation and vulnerability testing using DeepSeek API\ntriggers:\n  - help me test web vulnerabilities with deepseek pentest ai\n  - how do i use the deepseek burp suite extension\n  - generate ai payloads for sql injection testing\n  - automate fuzzing with deepseek pentest ai\n  - analyze web app parameters with burp ai extension\n  - export vulnerability findings from deepseek pentest\n  - send custom prompts to deepseek for payload generation\n  - configure deepseek api key in burp suite\n---\n\n# DeepSeek Pentest AI Burp Extension\n\n> Skill by [ara.so](https://ara.so) — Security Skills collection.\n\nDeepSeek Pentest AI is a Burp Suite extension that combines generative AI with intelligent fuzzing to automate payload generation and vulnerability testing. It uses the DeepSeek API to generate context-aware attack payloads for SQL injection, XSS, command injection, path traversal, SSRF, RCE, SSTI, XXE, and more.\n\n## What It Does\n\n- **AI-Powered Payload Generation**: Creates advanced attack payloads using DeepSeek's language model\n- **Automatic Parameter Detection**: Identifies parameters in GET, POST, JSON, XML, multipart, and custom headers\n- **Smart Fuzzing**: Injects payloads, compares against baselines, scores severity and confidence\n- **Real-Time Metrics**: Visualizes vulnerability types with integrated charts\n- **Privacy-First**: Redacts hostnames and sanitizes sensitive data before sending to AI\n- **Export Results**: CSV export with full request/response history and evidence\n- **Burp Integration**: Send findings directly to Repeater and Intruder\n\n## Installation\n\n### Prerequisites\n\n- Burp Suite Pro or Community Edition\n- Jython standalone JAR (for Python extensions)\n- DeepSeek API key from https://platform.deepseek.com/\n\n### Setup Steps\n\n1. **Clone the repository**:\n```bash\ngit clone https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI.git\ncd DeepSeek-Pentest-AI\n```\n\n2. **Configure Jython in Burp Suite**:\n   - Go to **Extender** → **Options**\n   - Under **Python Environment**, set the location of `jython-standalone.jar`\n\n3. **Load the extension**:\n   - Go to **Extender** → **Extensions** → **Add**\n   - Extension type: **Python**\n   - Extension file: Select the `.py` file from the cloned repository\n   - Check the **Output** tab for \"Plugin initialized\" message\n\n4. **Configure API Key**:\n   - Navigate to the **DeepSeek Pentest AI** tab in Burp\n   - Enter your DeepSeek API key (store in environment variable for security)\n\n## Core Workflow\n\n### 1. Capture a Request\n\nIntercept a request in Burp Proxy or send one from Repeater to the extension.\n\n### 2. Analyze & Generate Payloads\n\n```python\n# The extension automatically detects parameters from:\n# - Query strings: ?id=123&category=books\n# - POST body: username=admin&password=test\n# - JSON: {\"user\": \"admin\", \"role\": \"user\"}\n# - XML: <user><name>admin</name></user>\n# - Multipart form data\n# - Custom headers: X-Forwarded-For, User-Agent, etc.\n```\n\nIn the UI:\n- Select **Attack Type** (SQLi, XSS, Command Injection, etc.) or **CUSTOM PROMPT**\n- Set **Number of Payloads** (default: 10)\n- Set **Delay** between requests (milliseconds)\n- Click **Analyze & Generate**\n\n### 3. Start Fuzzing\n\nClick **Start Pentesting** to inject payloads into detected parameters and analyze responses.\n\n## Attack Types\n\nThe extension supports predefined attack strategies:\n\n```python\nattack_types = [\n    \"SQL Injection\",\n    \"XSS (Cross-Site Scripting)\",\n    \"Command Injection\",\n    \"Path Traversal\",\n    \"LFI (Local File Inclusion)\",\n    \"SSRF (Server-Side Request Forgery)\",\n    \"RCE (Remote Code Execution)\",\n    \"SSTI (Server-Side Template Injection)\",\n    \"XXE (XML External Entity)\",\n    \"NoSQL Injection\",\n    \"GraphQL Injection\",\n    \"Open Redirect\",\n    \"CRLF Injection\",\n    \"CORS Misconfiguration\",\n    \"Host Header Injection\",\n    \"CUSTOM PROMPT\"\n]\n```\n\n## Custom Prompt Usage\n\nFor specialized payload generation:\n\n```python\n# Instead of selecting a predefined attack type,\n# enter a custom instruction in the prompt field\n\n# Example 1: WAF Bypass\ncustom_prompt = \"Give me payloads SQLi boolean bypass WAF\"\n\n# Example 2: Specific Technology\ncustom_prompt = \"Generate SSTI payloads for Jinja2 templates\"\n\n# Example 3: Chained Attacks\ncustom_prompt = \"Create XSS payloads that also attempt DOM clobbering\"\n```\n\n**Result**: The AI generates targeted payloads matching your exact requirements instead of generic patterns.\n\n## Configuration\n\n### API Key Management\n\nStore your API key securely:\n\n```bash\n# Linux/macOS\nexport DEEPSEEK_API_KEY=\"your_api_key_here\"\n\n# Windows\nset DEEPSEEK_API_KEY=your_api_key_here\n```\n\nReference in code (if extending):\n```python\nimport os\napi_key = os.getenv('DEEPSEEK_API_KEY')\n```\n\n### Payload Generation Settings\n\n```python\n# Number of payloads to generate per attack type\nnum_payloads = 10  # Adjustable: 5-50\n\n# Delay between fuzzing requests (milliseconds)\ndelay_ms = 100  # Recommended: 100-1000 to avoid rate limits\n\n# Heuristic scoring thresholds\nconfidence_threshold = 0.7  # 0.0-1.0\nseverity_levels = [\"Low\", \"Medium\", \"High\", \"Critical\"]\n```\n\n## Code Examples\n\n### Example 1: Analyzing Generated Payloads\n\n```python\n# After clicking \"Analyze & Generate\", payloads appear in the AI Analysis tab\n# Example output for SQL Injection:\n\npayloads = [\n    \"' OR '1'='1\",\n    \"' OR '1'='1'--\",\n    \"admin' --\",\n    \"' OR 1=1--\",\n    \"' UNION SELECT NULL--\",\n    \"1' AND '1'='1\",\n    \"' OR 'a'='a\",\n    \"1' ORDER BY 1--\",\n    \"' OR ''='\",\n    \"1' UNION SELECT username, password FROM users--\"\n]\n\n# Each payload is tested against detected parameters\n# Results include: request, response, status code, length, evidence\n```\n\n### Example 2: Parameter Detection Logic\n\n```python\n# The extension automatically identifies injectable parameters:\n\n# Query string parameters\nGET /search?q=test&category=all HTTP/1.1\n# Detected: ['q', 'category']\n\n# JSON body\nPOST /api/login HTTP/1.1\nContent-Type: application/json\n\n{\"username\": \"admin\", \"password\": \"pass\"}\n# Detected: ['username', 'password']\n\n# XML body\nPOST /api/user HTTP/1.1\nContent-Type: application/xml\n\n<user><id>123</id><role>admin</role></user>\n# Detected: ['id', 'role']\n\n# Custom headers (special cases)\nGET / HTTP/1.1\nX-Forwarded-For: 127.0.0.1\nUser-Agent: Mozilla/5.0\n# Detected: ['X-Forwarded-For', 'User-Agent'] (if testing for header injection)\n```\n\n### Example 3: Heuristic Scoring\n\n```python\n# Response analysis after payload injection\n\ndef score_response(baseline, test_response, payload):\n    \"\"\"\n    Compares test response against baseline to detect anomalies\n    \"\"\"\n    score = 0.0\n    evidence = []\n    \n    # Check for SQL error messages\n    sql_errors = [\"SQL syntax\", \"mysql_fetch\", \"ORA-\", \"PostgreSQL\", \"sqlite3\"]\n    for error in sql_errors:\n        if error.lower() in test_response.lower():\n            score += 0.3\n            evidence.append(f\"SQL error detected: {error}\")\n    \n    # Check for XSS reflection\n    if payload in test_response and \"<script>\" in payload:\n        score += 0.4\n        evidence.append(\"XSS payload reflected in response\")\n    \n    # Check for status code changes\n    if test_response.status_code != baseline.status_code:\n        score += 0.1\n        evidence.append(f\"Status code changed: {baseline.status_code} -> {test_response.status_code}\")\n    \n    # Check for response length anomalies\n    if abs(len(test_response.body) - len(baseline.body)) > 500:\n        score += 0.2\n        evidence.append(\"Significant response length difference\")\n    \n    return min(score, 1.0), evidence\n```\n\n### Example 4: Exporting Results\n\n```python\n# Export options available in the UI:\n\n# 1. Export All Results\n# Includes: all requests, responses, payloads, parameters tested\n\n# 2. Export Only Vulnerabilities\n# Filters: only entries with confidence >= threshold\n\n# 3. Export With Evidence Only\n# Filters: entries with non-empty evidence snippets\n\n# CSV Format:\n# Timestamp | URL | Parameter | Payload | Status | Length | Confidence | Severity | Evidence\n```\n\n## Common Patterns\n\n### Pattern 1: Testing a Login Form\n\n1. Capture POST request to `/login`\n2. Extension detects `username` and `password` parameters\n3. Select **SQL Injection** attack type\n4. Generate 15 payloads\n5. Start fuzzing with 200ms delay\n6. Review results for authentication bypass evidence\n\n### Pattern 2: Custom WAF Bypass\n\n1. Capture request blocked by WAF\n2. Select **CUSTOM PROMPT**\n3. Enter: \"Generate SQLi payloads using URL encoding and inline comments to bypass ModSecurity\"\n4. Generate payloads\n5. Test manually in Repeater or auto-fuzz\n\n### Pattern 3: API Testing\n\n1. Capture JSON API request\n2. Extension auto-detects JSON parameters\n3. Select **NoSQL Injection** or **GraphQL Injection**\n4. Review AI-generated payloads for API-specific attacks\n5. Export findings with evidence\n\n## Troubleshooting\n\n### Extension Not Loading\n\n```bash\n# Check Extender → Output tab for errors\n\n# Common issues:\n# 1. Jython not configured correctly\n# Solution: Download jython-standalone.jar and set path in Extender → Options\n\n# 2. Python version mismatch\n# Solution: Use Jython 2.7.x (Burp requires Jython, not CPython)\n\n# 3. Missing dependencies\n# Solution: Ensure all imports are available in Jython environment\n```\n\n### API Key Errors\n\n```python\n# Error: \"Invalid API Key\" or \"Authentication failed\"\n\n# Solutions:\n# 1. Verify API key is correct from https://platform.deepseek.com/\n# 2. Check for extra spaces or newlines in key field\n# 3. Ensure API key has sufficient credits/quota\n# 4. Test API key with curl:\n\ncurl https://api.deepseek.com/v1/chat/completions \\\n  -H \"Content-Type: application/json\" \\\n  -H \"Authorization: Bearer $DEEPSEEK_API_KEY\" \\\n  -d '{\n    \"model\": \"deepseek-chat\",\n    \"messages\": [{\"role\": \"user\", \"content\": \"test\"}]\n  }'\n```\n\n### No Payloads Generated\n\n```python\n# Check AI Analysis tab for error messages\n\n# Common causes:\n# 1. API rate limit exceeded\n# Solution: Increase delay between requests\n\n# 2. Network connectivity issues\n# Solution: Check proxy settings, firewall rules\n\n# 3. Malformed request sent to AI\n# Solution: Review extension logs, check parameter detection\n\n# 4. AI returned unexpected format\n# Solution: Try different attack type or custom prompt\n```\n\n### False Positives\n\n```python\n# High confidence scores but manual verification shows no vulnerability\n\n# Mitigations:\n# 1. Adjust heuristic thresholds in code\n# 2. Review evidence snippets in Results tab\n# 3. Send to Repeater for manual confirmation\n# 4. Check baseline response accuracy\n# 5. Add custom scoring rules for your target\n```\n\n### Performance Issues\n\n```bash\n# Extension slows down Burp Suite\n\n# Solutions:\n# 1. Reduce number of payloads (5-10 instead of 20+)\n# 2. Increase delay between requests (500-1000ms)\n# 3. Disable real-time metrics if not needed\n# 4. Clear request history periodically\n# 5. Use \"Export & Clear\" to free memory\n```\n\n## Best Practices\n\n1. **Always test on authorized targets only** — Never use against systems without explicit permission\n2. **Start with baseline testing** — Let the extension capture a clean baseline response\n3. **Review AI-generated payloads** — Not all payloads may be relevant to your target\n4. **Use custom prompts for specific scenarios** — Generic attack types may miss edge cases\n5. **Verify findings manually** — AI scoring is heuristic, confirm vulnerabilities in Repeater\n6. **Export results regularly** — Prevents data loss and helps with reporting\n7. **Monitor API usage** — DeepSeek API has rate limits and costs\n8. **Sanitize exports** — Redact sensitive data before sharing CSV reports\n\n## Integration with Burp Tools\n\n### Send to Repeater\nRight-click any request in **Pentest Live** tab → **Send to Repeater** for manual testing\n\n### Send to Intruder\nRight-click any request → **Send to Intruder** → Use AI-generated payloads as position values\n\n### Scan Results\nCross-reference findings with Burp Scanner (Pro only) for comprehensive coverage\n\n## Environment Variables\n\n```bash\n# Recommended environment setup\nexport DEEPSEEK_API_KEY=\"sk-...\"\nexport BURP_PENTEST_DELAY=200\nexport BURP_PENTEST_PAYLOADS=10\nexport BURP_PENTEST_LOG_LEVEL=INFO\n```\n\n## Further Resources\n\n- Project Repository: https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI\n- DeepSeek API Docs: https://platform.deepseek.com/docs\n- Burp Extender API: https://portswigger.net/burp/extender/api/\n- Author LinkedIn: https://www.linkedin.com/in/hernanrodriguez-/\n"}, "c2f6995ca89080641475913b10365d2c1c9ccc47": {"t": 1785765411.8133883, "n": 12491, "h": "09509946e0143925", "c": "---\nname: deepseek-pentest-ai-burp-extension\ndescription: AI-powered Burp Suite extension for automated payload generation and vulnerability testing using DeepSeek API\ntriggers:\n  - help me test web vulnerabilities with deepseek pentest ai\n  - how do i use the deepseek burp suite extension\n  - generate ai payloads for sql injection testing\n  - automate fuzzing with deepseek pentest ai\n  - analyze web app parameters with burp ai extension\n  - export vulnerability findings from deepseek pentest\n  - send custom prompts to deepseek for payload generation\n  - configure deepseek api key in burp suite\n---\n\n# DeepSeek Pentest AI Burp Extension\n\n> Skill by [ara.so](https://ara.so) — Security Skills collection.\n\nDeepSeek Pentest AI is a Burp Suite extension that combines generative AI with intelligent fuzzing to automate payload generation and vulnerability testing. It uses the DeepSeek API to generate context-aware attack payloads for SQL injection, XSS, command injection, path traversal, SSRF, RCE, SSTI, XXE, and more.\n\n## What It Does\n\n- **AI-Powered Payload Generation**: Creates advanced attack payloads using DeepSeek's language model\n- **Automatic Parameter Detection**: Identifies parameters in GET, POST, JSON, XML, multipart, and custom headers\n- **Smart Fuzzing**: Injects payloads, compares against baselines, scores severity and confidence\n- **Real-Time Metrics**: Visualizes vulnerability types with integrated charts\n- **Privacy-First**: Redacts hostnames and sanitizes sensitive data before sending to AI\n- **Export Results**: CSV export with full request/response history and evidence\n- **Burp Integration**: Send findings directly to Repeater and Intruder\n\n## Installation\n\n### Prerequisites\n\n- Burp Suite Pro or Community Edition\n- Jython standalone JAR (for Python extensions)\n- DeepSeek API key from https://platform.deepseek.com/\n\n### Setup Steps\n\n1. **Clone the repository**:\n```bash\ngit clone https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI.git\ncd DeepSeek-Pentest-AI\n```\n\n2. **Configure Jython in Burp Suite**:\n   - Go to **Extender** → **Options**\n   - Under **Python Environment**, set the location of `jython-standalone.jar`\n\n3. **Load the extension**:\n   - Go to **Extender** → **Extensions** → **Add**\n   - Extension type: **Python**\n   - Extension file: Select the `.py` file from the cloned repository\n   - Check the **Output** tab for \"Plugin initialized\" message\n\n4. **Configure API Key**:\n   - Navigate to the **DeepSeek Pentest AI** tab in Burp\n   - Enter your DeepSeek API key (store in environment variable for security)\n\n## Core Workflow\n\n### 1. Capture a Request\n\nIntercept a request in Burp Proxy or send one from Repeater to the extension.\n\n### 2. Analyze & Generate Payloads\n\n```python\n# The extension automatically detects parameters from:\n# - Query strings: ?id=123&category=books\n# - POST body: username=admin&password=test\n# - JSON: {\"user\": \"admin\", \"role\": \"user\"}\n# - XML: <user><name>admin</name></user>\n# - Multipart form data\n# - Custom headers: X-Forwarded-For, User-Agent, etc.\n```\n\nIn the UI:\n- Select **Attack Type** (SQLi, XSS, Command Injection, etc.) or **CUSTOM PROMPT**\n- Set **Number of Payloads** (default: 10)\n- Set **Delay** between requests (milliseconds)\n- Click **Analyze & Generate**\n\n### 3. Start Fuzzing\n\nClick **Start Pentesting** to inject payloads into detected parameters and analyze responses.\n\n## Attack Types\n\nThe extension supports predefined attack strategies:\n\n```python\nattack_types = [\n    \"SQL Injection\",\n    \"XSS (Cross-Site Scripting)\",\n    \"Command Injection\",\n    \"Path Traversal\",\n    \"LFI (Local File Inclusion)\",\n    \"SSRF (Server-Side Request Forgery)\",\n    \"RCE (Remote Code Execution)\",\n    \"SSTI (Server-Side Template Injection)\",\n    \"XXE (XML External Entity)\",\n    \"NoSQL Injection\",\n    \"GraphQL Injection\",\n    \"Open Redirect\",\n    \"CRLF Injection\",\n    \"CORS Misconfiguration\",\n    \"Host Header Injection\",\n    \"CUSTOM PROMPT\"\n]\n```\n\n## Custom Prompt Usage\n\nFor specialized payload generation:\n\n```python\n# Instead of selecting a predefined attack type,\n# enter a custom instruction in the prompt field\n\n# Example 1: WAF Bypass\ncustom_prompt = \"Give me payloads SQLi boolean bypass WAF\"\n\n# Example 2: Specific Technology\ncustom_prompt = \"Generate SSTI payloads for Jinja2 templates\"\n\n# Example 3: Chained Attacks\ncustom_prompt = \"Create XSS payloads that also attempt DOM clobbering\"\n```\n\n**Result**: The AI generates targeted payloads matching your exact requirements instead of generic patterns.\n\n## Configuration\n\n### API Key Management\n\nStore your API key securely:\n\n```bash\n# Linux/macOS\nexport DEEPSEEK_API_KEY=\"your_api_key_here\"\n\n# Windows\nset DEEPSEEK_API_KEY=your_api_key_here\n```\n\nReference in code (if extending):\n```python\nimport os\napi_key = os.getenv('DEEPSEEK_API_KEY')\n```\n\n### Payload Generation Settings\n\n```python\n# Number of payloads to generate per attack type\nnum_payloads = 10  # Adjustable: 5-50\n\n# Delay between fuzzing requests (milliseconds)\ndelay_ms = 100  # Recommended: 100-1000 to avoid rate limits\n\n# Heuristic scoring thresholds\nconfidence_threshold = 0.7  # 0.0-1.0\nseverity_levels = [\"Low\", \"Medium\", \"High\", \"Critical\"]\n```\n\n## Code Examples\n\n### Example 1: Analyzing Generated Payloads\n\n```python\n# After clicking \"Analyze & Generate\", payloads appear in the AI Analysis tab\n# Example output for SQL Injection:\n\npayloads = [\n    \"' OR '1'='1\",\n    \"' OR '1'='1'--\",\n    \"admin' --\",\n    \"' OR 1=1--\",\n    \"' UNION SELECT NULL--\",\n    \"1' AND '1'='1\",\n    \"' OR 'a'='a\",\n    \"1' ORDER BY 1--\",\n    \"' OR ''='\",\n    \"1' UNION SELECT username, password FROM users--\"\n]\n\n# Each payload is tested against detected parameters\n# Results include: request, response, status code, length, evidence\n```\n\n### Example 2: Parameter Detection Logic\n\n```python\n# The extension automatically identifies injectable parameters:\n\n# Query string parameters\nGET /search?q=test&category=all HTTP/1.1\n# Detected: ['q', 'category']\n\n# JSON body\nPOST /api/login HTTP/1.1\nContent-Type: application/json\n\n{\"username\": \"admin\", \"password\": \"pass\"}\n# Detected: ['username', 'password']\n\n# XML body\nPOST /api/user HTTP/1.1\nContent-Type: application/xml\n\n<user><id>123</id><role>admin</role></user>\n# Detected: ['id', 'role']\n\n# Custom headers (special cases)\nGET / HTTP/1.1\nX-Forwarded-For: 127.0.0.1\nUser-Agent: Mozilla/5.0\n# Detected: ['X-Forwarded-For', 'User-Agent'] (if testing for header injection)\n```\n\n### Example 3: Heuristic Scoring\n\n```python\n# Response analysis after payload injection\n\ndef score_response(baseline, test_response, payload):\n    \"\"\"\n    Compares test response against baseline to detect anomalies\n    \"\"\"\n    score = 0.0\n    evidence = []\n    \n    # Check for SQL error messages\n    sql_errors = [\"SQL syntax\", \"mysql_fetch\", \"ORA-\", \"PostgreSQL\", \"sqlite3\"]\n    for error in sql_errors:\n        if error.lower() in test_response.lower():\n            score += 0.3\n            evidence.append(f\"SQL error detected: {error}\")\n    \n    # Check for XSS reflection\n    if payload in test_response and \"<script>\" in payload:\n        score += 0.4\n        evidence.append(\"XSS payload reflected in response\")\n    \n    # Check for status code changes\n    if test_response.status_code != baseline.status_code:\n        score += 0.1\n        evidence.append(f\"Status code changed: {baseline.status_code} -> {test_response.status_code}\")\n    \n    # Check for response length anomalies\n    if abs(len(test_response.body) - len(baseline.body)) > 500:\n        score += 0.2\n        evidence.append(\"Significant response length difference\")\n    \n    return min(score, 1.0), evidence\n```\n\n### Example 4: Exporting Results\n\n```python\n# Export options available in the UI:\n\n# 1. Export All Results\n# Includes: all requests, responses, payloads, parameters tested\n\n# 2. Export Only Vulnerabilities\n# Filters: only entries with confidence >= threshold\n\n# 3. Export With Evidence Only\n# Filters: entries with non-empty evidence snippets\n\n# CSV Format:\n# Timestamp | URL | Parameter | Payload | Status | Length | Confidence | Severity | Evidence\n```\n\n## Common Patterns\n\n### Pattern 1: Testing a Login Form\n\n1. Capture POST request to `/login`\n2. Extension detects `username` and `password` parameters\n3. Select **SQL Injection** attack type\n4. Generate 15 payloads\n5. Start fuzzing with 200ms delay\n6. Review results for authentication bypass evidence\n\n### Pattern 2: Custom WAF Bypass\n\n1. Capture request blocked by WAF\n2. Select **CUSTOM PROMPT**\n3. Enter: \"Generate SQLi payloads using URL encoding and inline comments to bypass ModSecurity\"\n4. Generate payloads\n5. Test manually in Repeater or auto-fuzz\n\n### Pattern 3: API Testing\n\n1. Capture JSON API request\n2. Extension auto-detects JSON parameters\n3. Select **NoSQL Injection** or **GraphQL Injection**\n4. Review AI-generated payloads for API-specific attacks\n5. Export findings with evidence\n\n## Troubleshooting\n\n### Extension Not Loading\n\n```bash\n# Check Extender → Output tab for errors\n\n# Common issues:\n# 1. Jython not configured correctly\n# Solution: Download jython-standalone.jar and set path in Extender → Options\n\n# 2. Python version mismatch\n# Solution: Use Jython 2.7.x (Burp requires Jython, not CPython)\n\n# 3. Missing dependencies\n# Solution: Ensure all imports are available in Jython environment\n```\n\n### API Key Errors\n\n```python\n# Error: \"Invalid API Key\" or \"Authentication failed\"\n\n# Solutions:\n# 1. Verify API key is correct from https://platform.deepseek.com/\n# 2. Check for extra spaces or newlines in key field\n# 3. Ensure API key has sufficient credits/quota\n# 4. Test API key with curl:\n\ncurl https://api.deepseek.com/v1/chat/completions \\\n  -H \"Content-Type: application/json\" \\\n  -H \"Authorization: Bearer $DEEPSEEK_API_KEY\" \\\n  -d '{\n    \"model\": \"deepseek-chat\",\n    \"messages\": [{\"role\": \"user\", \"content\": \"test\"}]\n  }'\n```\n\n### No Payloads Generated\n\n```python\n# Check AI Analysis tab for error messages\n\n# Common causes:\n# 1. API rate limit exceeded\n# Solution: Increase delay between requests\n\n# 2. Network connectivity issues\n# Solution: Check proxy settings, firewall rules\n\n# 3. Malformed request sent to AI\n# Solution: Review extension logs, check parameter detection\n\n# 4. AI returned unexpected format\n# Solution: Try different attack type or custom prompt\n```\n\n### False Positives\n\n```python\n# High confidence scores but manual verification shows no vulnerability\n\n# Mitigations:\n# 1. Adjust heuristic thresholds in code\n# 2. Review evidence snippets in Results tab\n# 3. Send to Repeater for manual confirmation\n# 4. Check baseline response accuracy\n# 5. Add custom scoring rules for your target\n```\n\n### Performance Issues\n\n```bash\n# Extension slows down Burp Suite\n\n# Solutions:\n# 1. Reduce number of payloads (5-10 instead of 20+)\n# 2. Increase delay between requests (500-1000ms)\n# 3. Disable real-time metrics if not needed\n# 4. Clear request history periodically\n# 5. Use \"Export & Clear\" to free memory\n```\n\n## Best Practices\n\n1. **Always test on authorized targets only** — Never use against systems without explicit permission\n2. **Start with baseline testing** — Let the extension capture a clean baseline response\n3. **Review AI-generated payloads** — Not all payloads may be relevant to your target\n4. **Use custom prompts for specific scenarios** — Generic attack types may miss edge cases\n5. **Verify findings manually** — AI scoring is heuristic, confirm vulnerabilities in Repeater\n6. **Export results regularly** — Prevents data loss and helps with reporting\n7. **Monitor API usage** — DeepSeek API has rate limits and costs\n8. **Sanitize exports** — Redact sensitive data before sharing CSV reports\n\n## Integration with Burp Tools\n\n### Send to Repeater\nRight-click any request in **Pentest Live** tab → **Send to Repeater** for manual testing\n\n### Send to Intruder\nRight-click any request → **Send to Intruder** → Use AI-generated payloads as position values\n\n### Scan Results\nCross-reference findings with Burp Scanner (Pro only) for comprehensive coverage\n\n## Environment Variables\n\n```bash\n# Recommended environment setup\nexport DEEPSEEK_API_KEY=\"sk-...\"\nexport BURP_PENTEST_DELAY=200\nexport BURP_PENTEST_PAYLOADS=10\nexport BURP_PENTEST_LOG_LEVEL=INFO\n```\n\n## Further Resources\n\n- Project Repository: https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI\n- DeepSeek API Docs: https://platform.deepseek.com/docs\n- Burp Extender API: https://portswigger.net/burp/extender/api/\n- Author LinkedIn: https://www.linkedin.com/in/hernanrodriguez-/\n"}, "a9916463362a83d1af26173740672ea00910fad9": {"t": 1786067806.7328029, "n": 12491, "h": "09509946e0143925", "c": "---\nname: deepseek-pentest-ai-burp-extension\ndescription: AI-powered Burp Suite extension for automated payload generation and vulnerability testing using DeepSeek API\ntriggers:\n  - help me test web vulnerabilities with deepseek pentest ai\n  - how do i use the deepseek burp suite extension\n  - generate ai payloads for sql injection testing\n  - automate fuzzing with deepseek pentest ai\n  - analyze web app parameters with burp ai extension\n  - export vulnerability findings from deepseek pentest\n  - send custom prompts to deepseek for payload generation\n  - configure deepseek api key in burp suite\n---\n\n# DeepSeek Pentest AI Burp Extension\n\n> Skill by [ara.so](https://ara.so) — Security Skills collection.\n\nDeepSeek Pentest AI is a Burp Suite extension that combines generative AI with intelligent fuzzing to automate payload generation and vulnerability testing. It uses the DeepSeek API to generate context-aware attack payloads for SQL injection, XSS, command injection, path traversal, SSRF, RCE, SSTI, XXE, and more.\n\n## What It Does\n\n- **AI-Powered Payload Generation**: Creates advanced attack payloads using DeepSeek's language model\n- **Automatic Parameter Detection**: Identifies parameters in GET, POST, JSON, XML, multipart, and custom headers\n- **Smart Fuzzing**: Injects payloads, compares against baselines, scores severity and confidence\n- **Real-Time Metrics**: Visualizes vulnerability types with integrated charts\n- **Privacy-First**: Redacts hostnames and sanitizes sensitive data before sending to AI\n- **Export Results**: CSV export with full request/response history and evidence\n- **Burp Integration**: Send findings directly to Repeater and Intruder\n\n## Installation\n\n### Prerequisites\n\n- Burp Suite Pro or Community Edition\n- Jython standalone JAR (for Python extensions)\n- DeepSeek API key from https://platform.deepseek.com/\n\n### Setup Steps\n\n1. **Clone the repository**:\n```bash\ngit clone https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI.git\ncd DeepSeek-Pentest-AI\n```\n\n2. **Configure Jython in Burp Suite**:\n   - Go to **Extender** → **Options**\n   - Under **Python Environment**, set the location of `jython-standalone.jar`\n\n3. **Load the extension**:\n   - Go to **Extender** → **Extensions** → **Add**\n   - Extension type: **Python**\n   - Extension file: Select the `.py` file from the cloned repository\n   - Check the **Output** tab for \"Plugin initialized\" message\n\n4. **Configure API Key**:\n   - Navigate to the **DeepSeek Pentest AI** tab in Burp\n   - Enter your DeepSeek API key (store in environment variable for security)\n\n## Core Workflow\n\n### 1. Capture a Request\n\nIntercept a request in Burp Proxy or send one from Repeater to the extension.\n\n### 2. Analyze & Generate Payloads\n\n```python\n# The extension automatically detects parameters from:\n# - Query strings: ?id=123&category=books\n# - POST body: username=admin&password=test\n# - JSON: {\"user\": \"admin\", \"role\": \"user\"}\n# - XML: <user><name>admin</name></user>\n# - Multipart form data\n# - Custom headers: X-Forwarded-For, User-Agent, etc.\n```\n\nIn the UI:\n- Select **Attack Type** (SQLi, XSS, Command Injection, etc.) or **CUSTOM PROMPT**\n- Set **Number of Payloads** (default: 10)\n- Set **Delay** between requests (milliseconds)\n- Click **Analyze & Generate**\n\n### 3. Start Fuzzing\n\nClick **Start Pentesting** to inject payloads into detected parameters and analyze responses.\n\n## Attack Types\n\nThe extension supports predefined attack strategies:\n\n```python\nattack_types = [\n    \"SQL Injection\",\n    \"XSS (Cross-Site Scripting)\",\n    \"Command Injection\",\n    \"Path Traversal\",\n    \"LFI (Local File Inclusion)\",\n    \"SSRF (Server-Side Request Forgery)\",\n    \"RCE (Remote Code Execution)\",\n    \"SSTI (Server-Side Template Injection)\",\n    \"XXE (XML External Entity)\",\n    \"NoSQL Injection\",\n    \"GraphQL Injection\",\n    \"Open Redirect\",\n    \"CRLF Injection\",\n    \"CORS Misconfiguration\",\n    \"Host Header Injection\",\n    \"CUSTOM PROMPT\"\n]\n```\n\n## Custom Prompt Usage\n\nFor specialized payload generation:\n\n```python\n# Instead of selecting a predefined attack type,\n# enter a custom instruction in the prompt field\n\n# Example 1: WAF Bypass\ncustom_prompt = \"Give me payloads SQLi boolean bypass WAF\"\n\n# Example 2: Specific Technology\ncustom_prompt = \"Generate SSTI payloads for Jinja2 templates\"\n\n# Example 3: Chained Attacks\ncustom_prompt = \"Create XSS payloads that also attempt DOM clobbering\"\n```\n\n**Result**: The AI generates targeted payloads matching your exact requirements instead of generic patterns.\n\n## Configuration\n\n### API Key Management\n\nStore your API key securely:\n\n```bash\n# Linux/macOS\nexport DEEPSEEK_API_KEY=\"your_api_key_here\"\n\n# Windows\nset DEEPSEEK_API_KEY=your_api_key_here\n```\n\nReference in code (if extending):\n```python\nimport os\napi_key = os.getenv('DEEPSEEK_API_KEY')\n```\n\n### Payload Generation Settings\n\n```python\n# Number of payloads to generate per attack type\nnum_payloads = 10  # Adjustable: 5-50\n\n# Delay between fuzzing requests (milliseconds)\ndelay_ms = 100  # Recommended: 100-1000 to avoid rate limits\n\n# Heuristic scoring thresholds\nconfidence_threshold = 0.7  # 0.0-1.0\nseverity_levels = [\"Low\", \"Medium\", \"High\", \"Critical\"]\n```\n\n## Code Examples\n\n### Example 1: Analyzing Generated Payloads\n\n```python\n# After clicking \"Analyze & Generate\", payloads appear in the AI Analysis tab\n# Example output for SQL Injection:\n\npayloads = [\n    \"' OR '1'='1\",\n    \"' OR '1'='1'--\",\n    \"admin' --\",\n    \"' OR 1=1--\",\n    \"' UNION SELECT NULL--\",\n    \"1' AND '1'='1\",\n    \"' OR 'a'='a\",\n    \"1' ORDER BY 1--\",\n    \"' OR ''='\",\n    \"1' UNION SELECT username, password FROM users--\"\n]\n\n# Each payload is tested against detected parameters\n# Results include: request, response, status code, length, evidence\n```\n\n### Example 2: Parameter Detection Logic\n\n```python\n# The extension automatically identifies injectable parameters:\n\n# Query string parameters\nGET /search?q=test&category=all HTTP/1.1\n# Detected: ['q', 'category']\n\n# JSON body\nPOST /api/login HTTP/1.1\nContent-Type: application/json\n\n{\"username\": \"admin\", \"password\": \"pass\"}\n# Detected: ['username', 'password']\n\n# XML body\nPOST /api/user HTTP/1.1\nContent-Type: application/xml\n\n<user><id>123</id><role>admin</role></user>\n# Detected: ['id', 'role']\n\n# Custom headers (special cases)\nGET / HTTP/1.1\nX-Forwarded-For: 127.0.0.1\nUser-Agent: Mozilla/5.0\n# Detected: ['X-Forwarded-For', 'User-Agent'] (if testing for header injection)\n```\n\n### Example 3: Heuristic Scoring\n\n```python\n# Response analysis after payload injection\n\ndef score_response(baseline, test_response, payload):\n    \"\"\"\n    Compares test response against baseline to detect anomalies\n    \"\"\"\n    score = 0.0\n    evidence = []\n    \n    # Check for SQL error messages\n    sql_errors = [\"SQL syntax\", \"mysql_fetch\", \"ORA-\", \"PostgreSQL\", \"sqlite3\"]\n    for error in sql_errors:\n        if error.lower() in test_response.lower():\n            score += 0.3\n            evidence.append(f\"SQL error detected: {error}\")\n    \n    # Check for XSS reflection\n    if payload in test_response and \"<script>\" in payload:\n        score += 0.4\n        evidence.append(\"XSS payload reflected in response\")\n    \n    # Check for status code changes\n    if test_response.status_code != baseline.status_code:\n        score += 0.1\n        evidence.append(f\"Status code changed: {baseline.status_code} -> {test_response.status_code}\")\n    \n    # Check for response length anomalies\n    if abs(len(test_response.body) - len(baseline.body)) > 500:\n        score += 0.2\n        evidence.append(\"Significant response length difference\")\n    \n    return min(score, 1.0), evidence\n```\n\n### Example 4: Exporting Results\n\n```python\n# Export options available in the UI:\n\n# 1. Export All Results\n# Includes: all requests, responses, payloads, parameters tested\n\n# 2. Export Only Vulnerabilities\n# Filters: only entries with confidence >= threshold\n\n# 3. Export With Evidence Only\n# Filters: entries with non-empty evidence snippets\n\n# CSV Format:\n# Timestamp | URL | Parameter | Payload | Status | Length | Confidence | Severity | Evidence\n```\n\n## Common Patterns\n\n### Pattern 1: Testing a Login Form\n\n1. Capture POST request to `/login`\n2. Extension detects `username` and `password` parameters\n3. Select **SQL Injection** attack type\n4. Generate 15 payloads\n5. Start fuzzing with 200ms delay\n6. Review results for authentication bypass evidence\n\n### Pattern 2: Custom WAF Bypass\n\n1. Capture request blocked by WAF\n2. Select **CUSTOM PROMPT**\n3. Enter: \"Generate SQLi payloads using URL encoding and inline comments to bypass ModSecurity\"\n4. Generate payloads\n5. Test manually in Repeater or auto-fuzz\n\n### Pattern 3: API Testing\n\n1. Capture JSON API request\n2. Extension auto-detects JSON parameters\n3. Select **NoSQL Injection** or **GraphQL Injection**\n4. Review AI-generated payloads for API-specific attacks\n5. Export findings with evidence\n\n## Troubleshooting\n\n### Extension Not Loading\n\n```bash\n# Check Extender → Output tab for errors\n\n# Common issues:\n# 1. Jython not configured correctly\n# Solution: Download jython-standalone.jar and set path in Extender → Options\n\n# 2. Python version mismatch\n# Solution: Use Jython 2.7.x (Burp requires Jython, not CPython)\n\n# 3. Missing dependencies\n# Solution: Ensure all imports are available in Jython environment\n```\n\n### API Key Errors\n\n```python\n# Error: \"Invalid API Key\" or \"Authentication failed\"\n\n# Solutions:\n# 1. Verify API key is correct from https://platform.deepseek.com/\n# 2. Check for extra spaces or newlines in key field\n# 3. Ensure API key has sufficient credits/quota\n# 4. Test API key with curl:\n\ncurl https://api.deepseek.com/v1/chat/completions \\\n  -H \"Content-Type: application/json\" \\\n  -H \"Authorization: Bearer $DEEPSEEK_API_KEY\" \\\n  -d '{\n    \"model\": \"deepseek-chat\",\n    \"messages\": [{\"role\": \"user\", \"content\": \"test\"}]\n  }'\n```\n\n### No Payloads Generated\n\n```python\n# Check AI Analysis tab for error messages\n\n# Common causes:\n# 1. API rate limit exceeded\n# Solution: Increase delay between requests\n\n# 2. Network connectivity issues\n# Solution: Check proxy settings, firewall rules\n\n# 3. Malformed request sent to AI\n# Solution: Review extension logs, check parameter detection\n\n# 4. AI returned unexpected format\n# Solution: Try different attack type or custom prompt\n```\n\n### False Positives\n\n```python\n# High confidence scores but manual verification shows no vulnerability\n\n# Mitigations:\n# 1. Adjust heuristic thresholds in code\n# 2. Review evidence snippets in Results tab\n# 3. Send to Repeater for manual confirmation\n# 4. Check baseline response accuracy\n# 5. Add custom scoring rules for your target\n```\n\n### Performance Issues\n\n```bash\n# Extension slows down Burp Suite\n\n# Solutions:\n# 1. Reduce number of payloads (5-10 instead of 20+)\n# 2. Increase delay between requests (500-1000ms)\n# 3. Disable real-time metrics if not needed\n# 4. Clear request history periodically\n# 5. Use \"Export & Clear\" to free memory\n```\n\n## Best Practices\n\n1. **Always test on authorized targets only** — Never use against systems without explicit permission\n2. **Start with baseline testing** — Let the extension capture a clean baseline response\n3. **Review AI-generated payloads** — Not all payloads may be relevant to your target\n4. **Use custom prompts for specific scenarios** — Generic attack types may miss edge cases\n5. **Verify findings manually** — AI scoring is heuristic, confirm vulnerabilities in Repeater\n6. **Export results regularly** — Prevents data loss and helps with reporting\n7. **Monitor API usage** — DeepSeek API has rate limits and costs\n8. **Sanitize exports** — Redact sensitive data before sharing CSV reports\n\n## Integration with Burp Tools\n\n### Send to Repeater\nRight-click any request in **Pentest Live** tab → **Send to Repeater** for manual testing\n\n### Send to Intruder\nRight-click any request → **Send to Intruder** → Use AI-generated payloads as position values\n\n### Scan Results\nCross-reference findings with Burp Scanner (Pro only) for comprehensive coverage\n\n## Environment Variables\n\n```bash\n# Recommended environment setup\nexport DEEPSEEK_API_KEY=\"sk-...\"\nexport BURP_PENTEST_DELAY=200\nexport BURP_PENTEST_PAYLOADS=10\nexport BURP_PENTEST_LOG_LEVEL=INFO\n```\n\n## Further Resources\n\n- Project Repository: https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI\n- DeepSeek API Docs: https://platform.deepseek.com/docs\n- Burp Extender API: https://portswigger.net/burp/extender/api/\n- Author LinkedIn: https://www.linkedin.com/in/hernanrodriguez-/\n"}, "304c245fe99201a24531b3cbb4afc3e695060fcf": {"t": 1786068468.1762269, "n": 12491, "h": "09509946e0143925", "c": "---\nname: deepseek-pentest-ai-burp-extension\ndescription: AI-powered Burp Suite extension for automated payload generation and vulnerability testing using DeepSeek API\ntriggers:\n  - help me test web vulnerabilities with deepseek pentest ai\n  - how do i use the deepseek burp suite extension\n  - generate ai payloads for sql injection testing\n  - automate fuzzing with deepseek pentest ai\n  - analyze web app parameters with burp ai extension\n  - export vulnerability findings from deepseek pentest\n  - send custom prompts to deepseek for payload generation\n  - configure deepseek api key in burp suite\n---\n\n# DeepSeek Pentest AI Burp Extension\n\n> Skill by [ara.so](https://ara.so) — Security Skills collection.\n\nDeepSeek Pentest AI is a Burp Suite extension that combines generative AI with intelligent fuzzing to automate payload generation and vulnerability testing. It uses the DeepSeek API to generate context-aware attack payloads for SQL injection, XSS, command injection, path traversal, SSRF, RCE, SSTI, XXE, and more.\n\n## What It Does\n\n- **AI-Powered Payload Generation**: Creates advanced attack payloads using DeepSeek's language model\n- **Automatic Parameter Detection**: Identifies parameters in GET, POST, JSON, XML, multipart, and custom headers\n- **Smart Fuzzing**: Injects payloads, compares against baselines, scores severity and confidence\n- **Real-Time Metrics**: Visualizes vulnerability types with integrated charts\n- **Privacy-First**: Redacts hostnames and sanitizes sensitive data before sending to AI\n- **Export Results**: CSV export with full request/response history and evidence\n- **Burp Integration**: Send findings directly to Repeater and Intruder\n\n## Installation\n\n### Prerequisites\n\n- Burp Suite Pro or Community Edition\n- Jython standalone JAR (for Python extensions)\n- DeepSeek API key from https://platform.deepseek.com/\n\n### Setup Steps\n\n1. **Clone the repository**:\n```bash\ngit clone https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI.git\ncd DeepSeek-Pentest-AI\n```\n\n2. **Configure Jython in Burp Suite**:\n   - Go to **Extender** → **Options**\n   - Under **Python Environment**, set the location of `jython-standalone.jar`\n\n3. **Load the extension**:\n   - Go to **Extender** → **Extensions** → **Add**\n   - Extension type: **Python**\n   - Extension file: Select the `.py` file from the cloned repository\n   - Check the **Output** tab for \"Plugin initialized\" message\n\n4. **Configure API Key**:\n   - Navigate to the **DeepSeek Pentest AI** tab in Burp\n   - Enter your DeepSeek API key (store in environment variable for security)\n\n## Core Workflow\n\n### 1. Capture a Request\n\nIntercept a request in Burp Proxy or send one from Repeater to the extension.\n\n### 2. Analyze & Generate Payloads\n\n```python\n# The extension automatically detects parameters from:\n# - Query strings: ?id=123&category=books\n# - POST body: username=admin&password=test\n# - JSON: {\"user\": \"admin\", \"role\": \"user\"}\n# - XML: <user><name>admin</name></user>\n# - Multipart form data\n# - Custom headers: X-Forwarded-For, User-Agent, etc.\n```\n\nIn the UI:\n- Select **Attack Type** (SQLi, XSS, Command Injection, etc.) or **CUSTOM PROMPT**\n- Set **Number of Payloads** (default: 10)\n- Set **Delay** between requests (milliseconds)\n- Click **Analyze & Generate**\n\n### 3. Start Fuzzing\n\nClick **Start Pentesting** to inject payloads into detected parameters and analyze responses.\n\n## Attack Types\n\nThe extension supports predefined attack strategies:\n\n```python\nattack_types = [\n    \"SQL Injection\",\n    \"XSS (Cross-Site Scripting)\",\n    \"Command Injection\",\n    \"Path Traversal\",\n    \"LFI (Local File Inclusion)\",\n    \"SSRF (Server-Side Request Forgery)\",\n    \"RCE (Remote Code Execution)\",\n    \"SSTI (Server-Side Template Injection)\",\n    \"XXE (XML External Entity)\",\n    \"NoSQL Injection\",\n    \"GraphQL Injection\",\n    \"Open Redirect\",\n    \"CRLF Injection\",\n    \"CORS Misconfiguration\",\n    \"Host Header Injection\",\n    \"CUSTOM PROMPT\"\n]\n```\n\n## Custom Prompt Usage\n\nFor specialized payload generation:\n\n```python\n# Instead of selecting a predefined attack type,\n# enter a custom instruction in the prompt field\n\n# Example 1: WAF Bypass\ncustom_prompt = \"Give me payloads SQLi boolean bypass WAF\"\n\n# Example 2: Specific Technology\ncustom_prompt = \"Generate SSTI payloads for Jinja2 templates\"\n\n# Example 3: Chained Attacks\ncustom_prompt = \"Create XSS payloads that also attempt DOM clobbering\"\n```\n\n**Result**: The AI generates targeted payloads matching your exact requirements instead of generic patterns.\n\n## Configuration\n\n### API Key Management\n\nStore your API key securely:\n\n```bash\n# Linux/macOS\nexport DEEPSEEK_API_KEY=\"your_api_key_here\"\n\n# Windows\nset DEEPSEEK_API_KEY=your_api_key_here\n```\n\nReference in code (if extending):\n```python\nimport os\napi_key = os.getenv('DEEPSEEK_API_KEY')\n```\n\n### Payload Generation Settings\n\n```python\n# Number of payloads to generate per attack type\nnum_payloads = 10  # Adjustable: 5-50\n\n# Delay between fuzzing requests (milliseconds)\ndelay_ms = 100  # Recommended: 100-1000 to avoid rate limits\n\n# Heuristic scoring thresholds\nconfidence_threshold = 0.7  # 0.0-1.0\nseverity_levels = [\"Low\", \"Medium\", \"High\", \"Critical\"]\n```\n\n## Code Examples\n\n### Example 1: Analyzing Generated Payloads\n\n```python\n# After clicking \"Analyze & Generate\", payloads appear in the AI Analysis tab\n# Example output for SQL Injection:\n\npayloads = [\n    \"' OR '1'='1\",\n    \"' OR '1'='1'--\",\n    \"admin' --\",\n    \"' OR 1=1--\",\n    \"' UNION SELECT NULL--\",\n    \"1' AND '1'='1\",\n    \"' OR 'a'='a\",\n    \"1' ORDER BY 1--\",\n    \"' OR ''='\",\n    \"1' UNION SELECT username, password FROM users--\"\n]\n\n# Each payload is tested against detected parameters\n# Results include: request, response, status code, length, evidence\n```\n\n### Example 2: Parameter Detection Logic\n\n```python\n# The extension automatically identifies injectable parameters:\n\n# Query string parameters\nGET /search?q=test&category=all HTTP/1.1\n# Detected: ['q', 'category']\n\n# JSON body\nPOST /api/login HTTP/1.1\nContent-Type: application/json\n\n{\"username\": \"admin\", \"password\": \"pass\"}\n# Detected: ['username', 'password']\n\n# XML body\nPOST /api/user HTTP/1.1\nContent-Type: application/xml\n\n<user><id>123</id><role>admin</role></user>\n# Detected: ['id', 'role']\n\n# Custom headers (special cases)\nGET / HTTP/1.1\nX-Forwarded-For: 127.0.0.1\nUser-Agent: Mozilla/5.0\n# Detected: ['X-Forwarded-For', 'User-Agent'] (if testing for header injection)\n```\n\n### Example 3: Heuristic Scoring\n\n```python\n# Response analysis after payload injection\n\ndef score_response(baseline, test_response, payload):\n    \"\"\"\n    Compares test response against baseline to detect anomalies\n    \"\"\"\n    score = 0.0\n    evidence = []\n    \n    # Check for SQL error messages\n    sql_errors = [\"SQL syntax\", \"mysql_fetch\", \"ORA-\", \"PostgreSQL\", \"sqlite3\"]\n    for error in sql_errors:\n        if error.lower() in test_response.lower():\n            score += 0.3\n            evidence.append(f\"SQL error detected: {error}\")\n    \n    # Check for XSS reflection\n    if payload in test_response and \"<script>\" in payload:\n        score += 0.4\n        evidence.append(\"XSS payload reflected in response\")\n    \n    # Check for status code changes\n    if test_response.status_code != baseline.status_code:\n        score += 0.1\n        evidence.append(f\"Status code changed: {baseline.status_code} -> {test_response.status_code}\")\n    \n    # Check for response length anomalies\n    if abs(len(test_response.body) - len(baseline.body)) > 500:\n        score += 0.2\n        evidence.append(\"Significant response length difference\")\n    \n    return min(score, 1.0), evidence\n```\n\n### Example 4: Exporting Results\n\n```python\n# Export options available in the UI:\n\n# 1. Export All Results\n# Includes: all requests, responses, payloads, parameters tested\n\n# 2. Export Only Vulnerabilities\n# Filters: only entries with confidence >= threshold\n\n# 3. Export With Evidence Only\n# Filters: entries with non-empty evidence snippets\n\n# CSV Format:\n# Timestamp | URL | Parameter | Payload | Status | Length | Confidence | Severity | Evidence\n```\n\n## Common Patterns\n\n### Pattern 1: Testing a Login Form\n\n1. Capture POST request to `/login`\n2. Extension detects `username` and `password` parameters\n3. Select **SQL Injection** attack type\n4. Generate 15 payloads\n5. Start fuzzing with 200ms delay\n6. Review results for authentication bypass evidence\n\n### Pattern 2: Custom WAF Bypass\n\n1. Capture request blocked by WAF\n2. Select **CUSTOM PROMPT**\n3. Enter: \"Generate SQLi payloads using URL encoding and inline comments to bypass ModSecurity\"\n4. Generate payloads\n5. Test manually in Repeater or auto-fuzz\n\n### Pattern 3: API Testing\n\n1. Capture JSON API request\n2. Extension auto-detects JSON parameters\n3. Select **NoSQL Injection** or **GraphQL Injection**\n4. Review AI-generated payloads for API-specific attacks\n5. Export findings with evidence\n\n## Troubleshooting\n\n### Extension Not Loading\n\n```bash\n# Check Extender → Output tab for errors\n\n# Common issues:\n# 1. Jython not configured correctly\n# Solution: Download jython-standalone.jar and set path in Extender → Options\n\n# 2. Python version mismatch\n# Solution: Use Jython 2.7.x (Burp requires Jython, not CPython)\n\n# 3. Missing dependencies\n# Solution: Ensure all imports are available in Jython environment\n```\n\n### API Key Errors\n\n```python\n# Error: \"Invalid API Key\" or \"Authentication failed\"\n\n# Solutions:\n# 1. Verify API key is correct from https://platform.deepseek.com/\n# 2. Check for extra spaces or newlines in key field\n# 3. Ensure API key has sufficient credits/quota\n# 4. Test API key with curl:\n\ncurl https://api.deepseek.com/v1/chat/completions \\\n  -H \"Content-Type: application/json\" \\\n  -H \"Authorization: Bearer $DEEPSEEK_API_KEY\" \\\n  -d '{\n    \"model\": \"deepseek-chat\",\n    \"messages\": [{\"role\": \"user\", \"content\": \"test\"}]\n  }'\n```\n\n### No Payloads Generated\n\n```python\n# Check AI Analysis tab for error messages\n\n# Common causes:\n# 1. API rate limit exceeded\n# Solution: Increase delay between requests\n\n# 2. Network connectivity issues\n# Solution: Check proxy settings, firewall rules\n\n# 3. Malformed request sent to AI\n# Solution: Review extension logs, check parameter detection\n\n# 4. AI returned unexpected format\n# Solution: Try different attack type or custom prompt\n```\n\n### False Positives\n\n```python\n# High confidence scores but manual verification shows no vulnerability\n\n# Mitigations:\n# 1. Adjust heuristic thresholds in code\n# 2. Review evidence snippets in Results tab\n# 3. Send to Repeater for manual confirmation\n# 4. Check baseline response accuracy\n# 5. Add custom scoring rules for your target\n```\n\n### Performance Issues\n\n```bash\n# Extension slows down Burp Suite\n\n# Solutions:\n# 1. Reduce number of payloads (5-10 instead of 20+)\n# 2. Increase delay between requests (500-1000ms)\n# 3. Disable real-time metrics if not needed\n# 4. Clear request history periodically\n# 5. Use \"Export & Clear\" to free memory\n```\n\n## Best Practices\n\n1. **Always test on authorized targets only** — Never use against systems without explicit permission\n2. **Start with baseline testing** — Let the extension capture a clean baseline response\n3. **Review AI-generated payloads** — Not all payloads may be relevant to your target\n4. **Use custom prompts for specific scenarios** — Generic attack types may miss edge cases\n5. **Verify findings manually** — AI scoring is heuristic, confirm vulnerabilities in Repeater\n6. **Export results regularly** — Prevents data loss and helps with reporting\n7. **Monitor API usage** — DeepSeek API has rate limits and costs\n8. **Sanitize exports** — Redact sensitive data before sharing CSV reports\n\n## Integration with Burp Tools\n\n### Send to Repeater\nRight-click any request in **Pentest Live** tab → **Send to Repeater** for manual testing\n\n### Send to Intruder\nRight-click any request → **Send to Intruder** → Use AI-generated payloads as position values\n\n### Scan Results\nCross-reference findings with Burp Scanner (Pro only) for comprehensive coverage\n\n## Environment Variables\n\n```bash\n# Recommended environment setup\nexport DEEPSEEK_API_KEY=\"sk-...\"\nexport BURP_PENTEST_DELAY=200\nexport BURP_PENTEST_PAYLOADS=10\nexport BURP_PENTEST_LOG_LEVEL=INFO\n```\n\n## Further Resources\n\n- Project Repository: https://github.com/HernanRodriguez1/DeepSeek-Pentest-AI\n- DeepSeek API Docs: https://platform.deepseek.com/docs\n- Burp Extender API: https://portswigger.net/burp/extender/api/\n- Author LinkedIn: https://www.linkedin.com/in/hernanrodriguez-/\n"}}}}