# DarkMoon

An open-source autonomous AI penetration-testing tool that runs specialist agents across web, API, cloud, Active Directory, Kubernetes and CI/CD targets, documenting each finding with the command and raw output that produced it.

DarkMoon is autonomous AI penetration testing for infrastructure the operator is authorized to test: pointed at a target, it runs a full assessment with 50 specialist agents covering web, API, cloud, Active Directory, Kubernetes, CI/CD and AI/LLM surfaces mapped to the OWASP LLM Top 10, then documents every finding with the exact command and raw output that produced it.

The open-source edition is a command-line tool, GPLv3 licensed and self-hosted, with every agent's methodology written as plain Markdown that a user can read, diff or fork. It runs on a local LLM behind a Privacy Gateway that tokenizes real IPs, hosts and credentials, so the model reasons over placeholders while real values stay on the user's own perimeter.

It integrates with GitHub, GitLab, Jenkins, VS Code, JetBrains, n8n, Grafana and Splunk, though the CLI-based subset works with the open-source edition; a separate paid edition adds a machine-verified remediation retest.

Website: https://dark-moon.org/