oss-fuzz-mcp

# OSS-Fuzz MCP Server This project provides a Model Context Protocol (MCP) server that wraps OSS-Fuzz's helper tools. It enables AI agents and other tools to interact with OSS-Fuzz through structured commands. ## Features - ✅ Project generation and management - ✅ Docker image building and management - ✅ Fuzzer compilation and execution - ✅ Code coverage analysis - ✅ Corpus management (GCS integration) - ✅ GitHub Actions integration (CIFuzz) ## Requirements 1. A local clone of the OSS-Fuzz repository 2. Python 3.7+ 3. Docker 4. gsutil (for corpus management) Install dependencies: ```bash pip install -r requirements.txt ``` ## File Structure | File | Purpose | |------|---------| | `server.py` | Main FastMCP server exposing OSS-Fuzz tools | | `requirements.txt` | Python package dependencies | ## Setup and Usage 1. Start the MCP server: ```bash python server.py ``` 2. Configure the OSS-Fuzz directory path: ```bash curl -X POST http://localhost:8000/tool/configure_oss_fuzz_path \ -H "Content-Type: application/json" \ -d '{"path": "/path/to/oss-fuzz"}' ``` 3. The server exposes various tools for: - Generating new projects - Building and running fuzzers - Managing corpora - Generating coverage reports - Setting up GitHub Actions integration