Bug Bounty MCP Server
Enables AI agents to generate and manage specialized bug bounty hunting workflows including reconnaissance, vulnerability testing, OSINT gathering, and file upload testing. Provides REST API endpoints for comprehensive security assessments with intelligence-driven vulnerability prioritization.
README
Bug Bounty MCP Server
A clean, focused server containing bug bounty hunting workflows and REST API endpoints.
Features
- Clean Architecture: Removed bloat and unnecessary dependencies while maintaining core functionality
- Bug Bounty Focused: Specialized workflows for reconnaissance, vulnerability hunting, business logic testing, OSINT, and file upload testing
- REST API Endpoints: Simple HTTP API for workflow generation and management
- Comprehensive Assessments: Combine multiple workflows for complete bug bounty assessments
Architecture
Core Components
- REST API Server (
src/server.py) - Flask-based HTTP API server with bug bounty workflow endpoints - MCP Server (
src/mcp_server.py) - FastMCP-based server for AI agent communication - Bug Bounty Workflows - Specialized workflow generation for different phases of testing
- Tool Integration - Comprehensive collection of security testing tools
Quick Start
1. Install Dependencies & Start the Server
# Install dependencies with uv
uv sync
# Install development dependencies (optional)
uv sync --dev
# Set up pre-commit hooks (recommended for development)
uv run pre-commit install
# Start the server
uv run src/server.py --debug --port 8888
# Or use the launcher script
./start-server.sh --debug
2. Test the API
# Create reconnaissance workflow
curl -X POST http://127.0.0.1:8888/api/bugbounty/reconnaissance-workflow \
-H "Content-Type: application/json" \
-d '{"domain": "example.com", "program_type": "web"}'
Configuration
Environment Variables
BUGBOUNTY_MCP_PORT: Server port (default: 8888)BUGBOUNTY_MCP_HOST: Server host (default: 127.0.0.1)DEBUG: Enable debug mode (default: false)
Command Line Options
uv run src/server.py --help
--debug: Enable debug mode--port PORT: Set server port--host HOST: Set server host
Key Features
- Bug Bounty Workflow Management: Complete workflow generation for different phases of bug bounty hunting
- Vulnerability Prioritization: Intelligence-driven prioritization based on impact and bounty potential
- File Upload Testing: Specialized framework for file upload vulnerability testing
- OSINT Integration: Comprehensive OSINT gathering workflows
- Business Logic Testing: Structured approach to business logic vulnerability discovery
Dependencies
Project uses uv for fast, reliable dependency management:
Core Dependencies
- Flask: Web framework for REST API
- FastMCP: MCP server framework
- Requests: HTTP client library
- Python 3.11+: Core runtime (supports Python 3.11, 3.12, 3.13)
Development Dependencies
- Ruff: Fast Python linter and formatter
- Bandit: Security vulnerability scanner
- Pydocstyle: Documentation quality checker
- Pyright: Static type checker
- Pre-commit: Git pre-commit hooks framework
Install dependencies:
uv sync # Core dependencies only
uv sync --dev # Include development tools
Add new dependencies:
uv add package-name
Code Quality
This project enforces code quality through automated pre-commit hooks:
# Install pre-commit hooks
uv run pre-commit install
# Run checks on all files
uv run pre-commit run --all-files
# Run specific checks
uv run ruff check # Linting
uv run ruff format # Formatting
uv run bandit -c pyproject.toml # Security scan
uv run pydocstyle # Documentation check
Standards:
- Line length: 88 characters
- Documentation: Google docstring convention
- Type hints: Required for public APIs
- Security: Bandit security scanning enabled
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。