An MCP server that connects LLMs to 150+ cybersecurity tools for autonomous penetration testing and vulnerability discovery. It coordinates 12+ specialized agents to execute offensive security workflows against target domains or IP addresses.
Project overview
Bridges a documented arsenal of 150+ professional security tools directly to LLMs via MCP, enabling AI agents to autonomously run penetration testing tasks.
Project type
MCP · AI Agent
Deployment
Refer to project documentation
License
MIT
Best for
Developers, researchers, and operations teams running autonomous AI-agent automation for penetration testing and vulnerability discovery.
Key capabilities
Bridges LLMs with 150+ security tools to autonomously run penetration testing tasks.
Analyzes targets and selects optimal testing strategies.
Includes 12+ specialized AI agents for bug bounty, CTF solving, CVE intelligence, and exploit generation.
Provides headless Chrome automation for web testing with DOM analysis and network monitoring.
Enables live command control and monitoring of active processes.
Provides intelligent result caching with LRU eviction.
Provides a full-featured desktop client.
Limitations and risks
Requires specific prompting techniques to bypass LLM ethical constraints during operation.
AI agents can execute arbitrary security tools with powerful system access, necessitating strict operational isolation.
Getting started
Setup difficulty is rated medium due to the need to install multiple system dependencies and perform manual integration with AI clients. The documented first success path is to clone the repository, create a virtual environment, install requirements, start the server, and verify the health endpoint.