REDCELL

—

Autonomous LLM agents run an end-to-end pentest in a Kali container.

MIT

AI red-team platform where an orchestrator plans an engagement and delegates objectives to executor agents that run real tools inside a Kali container. Wraps nmap, nuclei, directory and vhost discovery and Metasploit as structured tools, handles reverse shells and pivoting through a caught shell, drives a Chromium browser for login flows and can scan public repos or local folders, with an operator console showing a live agent graph and findings triage. A LangGraph plan/act loop reaches models through LiteLLM, and reports export to PDF, JSON and SARIF.

Does

Scan Exploit Automate

Web apps Networks

For

Red team Pentest