# PentestAgent

An AI agent framework for black-box penetration testing, supporting bug bounty, red-team and pentesting workflows from a terminal interface, with tools such as nmap, Metasploit and sqlmap.

PentestAgent is an AI agent framework for black-box security testing, supporting bug bounty, red-team and penetration-testing workflows from a terminal user interface. It works with any LiteLLM-supported model, including OpenAI and Anthropic models, and can be pointed at a custom OpenAI- or Anthropic-compatible relay endpoint through an environment variable.

It can run against a target directly from the terminal, or inside a Docker container for isolation with pre-installed pentesting tools: a base image bundling nmap, netcat and curl, or a Kali-based image additionally bundling Metasploit, sqlmap and Hydra, with the agent driving these tools directly through its terminal tool. It offers several modes accessible through in-app commands, including a single-shot assist mode for one instruction with tool execution.

It requires Python 3.10 or newer and Playwright's Chromium browser for its browser tool, and is released under the MIT license.

Website: https://github.com/GH05TCREW/pentestagent