va-pentest-mcp
A Model Context Protocol server for automated security vulnerability assessment, combining OWASP Dependency-Check dependency scanning with custom code vulnerability detection, and generating detailed HTML and JSON reports.
README
VA Pentest MCP - Vulnerability Assessment & Penetration Testing MCP Server
🔒 Model Context Protocol (MCP) Server for Security Vulnerability Assessment and Penetration Testing
A comprehensive Python-based MCP server that provides automated security scanning capabilities for projects. Combines OWASP Dependency-Check for dependency scanning with custom code vulnerability detection.
Features
🔍 Security Scanning
- Dependency Scanning: Uses OWASP Dependency-Check to identify known vulnerabilities in project dependencies
- Code Security Scanning: Custom regex-based vulnerability detection for:
- 🔐 Hardcoded Secrets (API keys, passwords, private keys, JWT tokens, database URLs, AWS credentials)
- 💉 SQL Injection patterns
- 🎯 Command Injection vulnerabilities
- ⚠️ Unsafe Operations (eval, pickle, dangerous imports, file permissions)
- 🌐 Cross-Site Scripting (XSS) vulnerabilities
- 🚶 Path Traversal vulnerabilities
📊 Report Generation
- HTML Reports: Beautiful, interactive HTML reports with vulnerability details and recommendations
- JSON Reports: Machine-readable JSON format for integration with other systems
- Comprehensive Summary: Risk assessment and vulnerability statistics
🛠️ MCP Protocol Support
- Full Model Context Protocol implementation
- Tool registration and management
- Resource handling
- Standardized request/response format
Installation
Prerequisites
- Python 3.8 or higher
- pip (Python package manager)
- OWASP Dependency-Check (optional - will attempt auto-install)
Setup
- Clone the repository
git clone https://github.com/banhongit7/va-pentest-mcp.git
cd va-pentest-mcp
- Create virtual environment
python -m venv venv
# On Windows
venv\Scripts\activate
# On macOS/Linux
source venv/bin/activate
- Install dependencies
pip install -r requirements.txt
- Install package
pip install -e .
Configuration
Environment Variables
Create a .env file in the project root:
# MCP Server Configuration
LOG_LEVEL=INFO
LOG_FILE=./logs/va-pentest-mcp.log
# Dependency Check Configuration
DEPENDENCY_CHECK_ENABLED=true
DEPENDENCY_CHECK_PATH=/path/to/dependency-check/bin/dependency-check
DEPENDENCY_CHECK_DB_DIR=./tools/dependency-check-db
# Code Scanner Configuration
CODE_SCANNER_ENABLED=true
SCAN_FOR_SECRETS=true
SCAN_FOR_SQL_INJECTION=true
SCAN_FOR_COMMAND_INJECTION=true
SCAN_FOR_UNSAFE_OPERATIONS=true
# Report Generation
GENERATE_HTML_REPORT=true
GENERATE_JSON_REPORT=true
# Python Path (for Windows)
PYTHONPATH=D:\mcp\va-pentest-mcp\src
MCP Configuration (config.json)
For OpenCode AI integration:
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"va-pentest-mcp": {
"type": "local",
"command": ["python", "-m", "va_pentest_mcp.server"],
"enabled": true,
"env": {
"PYTHONPATH": "D:\\mcp\\va-pentest-mcp\\src",
"PATH": "D:\\mcp\\va-pentest-mcp\\tools;{env:PATH}"
}
}
}
}
Usage
Starting the Server
python -m va_pentest_mcp.server
Or using the console script:
va-pentest-mcp
Available Tools
1. scan_dependencies
Scans project dependencies for known vulnerabilities using OWASP Dependency-Check.
{
"project_path": "/path/to/project"
}
2. scan_code
Scans source code for security vulnerabilities.
{
"project_path": "/path/to/project",
"file_extensions": [".py", ".js", ".java"] # Optional
}
3. scan_full
Runs complete security scan and generates reports.
{
"project_path": "/path/to/project",
"generate_reports": true
}
Project Structure
va-pentest-mcp/
├── src/
│ └── va_pentest_mcp/
│ ├── __init__.py # Package initialization
│ ├── server.py # Main MCP Server
│ ├── mcp_protocol.py # MCP Protocol Handler
│ ├── dependency_scanner.py # OWASP Dependency-Check Integration
│ ├── code_scanner.py # Custom Security Vulnerability Detection
│ ├── report_generator.py # HTML & JSON Report Generation
│ ├── config.py # Configuration Management
│ └── utils.py # Utility Functions
├── tools/ # External tools directory
│ ├── dependency-check/ # OWASP Dependency-Check binary
│ └── dependency-check-db/ # Vulnerability database
├── reports/ # Generated reports
│ ├── scan_report_YYYYMMDD_HHMMSS.html
│ └── scan_report_YYYYMMDD_HHMMSS.json
├── logs/ # Application logs
├── requirements.txt # Python dependencies
├── setup.py # Package setup script
├── README.md # This file
└── .gitignore # Git ignore rules
Vulnerability Detection
Hardcoded Secrets Detection
Detects:
- API Keys
- Passwords
- Private Keys (RSA, OPENSSH, DSA, EC)
- AWS Access Keys (AKIA...)
- JWT Tokens
- Database Connection Strings
Code Injection Detection
- SQL Injection via string concatenation, format strings, and direct queries
- Command Injection via os.system, subprocess, shell commands
Unsafe Operations
- eval() usage
- pickle/marshal usage
- Dangerous imports
- Insecure file permissions
Web Vulnerabilities
- XSS via innerHTML, dangerouslySetInnerHTML
- Path Traversal patterns
Report Output
HTML Report Features
- Executive Summary with Risk Assessment
- Vulnerability Statistics by Severity
- Detailed Vulnerability Listing
- Code Snippets with Line Numbers
- Remediation Recommendations
- Professional Styling
JSON Report Structure
{
"scan_info": {
"timestamp": "2024-01-01T12:00:00",
"project_path": "/path/to/project",
"scan_type": "Combined VA Scan"
},
"code_scan": {
"total_vulnerabilities": 5,
"vulnerabilities": [
{
"type": "Hardcoded Secret",
"severity": "CRITICAL",
"file": "config.py",
"line": 42,
"description": "Hardcoded API Key detected",
"recommendation": "Use environment variables instead"
}
]
},
"summary": {
"overall_risk": "HIGH",
"critical": 2,
"high": 3
}
}
Advanced Configuration
Custom File Extensions
file_extensions = ['.py', '.js', '.ts', '.java', '.go', '.rb', '.php', '.cs']
Severity Levels
- CRITICAL: Immediate action required
- HIGH: Significant security risk
- MEDIUM: Moderate risk, should be addressed
- LOW: Low risk, consider fixing
- INFO: Informational findings
Troubleshooting
dependency-check Not Found
# Try installing via pip
pip install dependency-check
# Or via npm
npm install -g dependency-check
# Or via homebrew (macOS)
brew install dependency-check
Permission Issues
# Make dependency-check executable
chmod +x /path/to/dependency-check
Windows Python Path
Ensure PYTHONPATH is correctly set in config.json:
"PYTHONPATH": "C:\\Users\\username\\va-pentest-mcp\\src"
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
MIT License - See LICENSE file for details
Security Note
This tool is designed for authorized security testing only. Always get proper authorization before testing security vulnerabilities in any system.
Support
For issues, questions, or suggestions:
- GitHub Issues: https://github.com/banhongit7/va-pentest-mcp/issues
- Email: banhongit7@gmail.com
VA Pentest MCP - Making security assessment automated and accessible 🚀
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。