#121 · Primary category: Cybersecurity & Decryption Tools

Dark-Moon

active-directory ai-agents ai-red-team ai-security-tool autonomous-agents cloud-security firmware-security iot-security kubernetes llm local-llm mcp multi-agent-systems offensive-security penetration-testing pentesting red-team security-automation security-tools self-hosted

Autonomous AI pentesting engine that runs full offensive security across web, cloud, AD, K8s, and IoT, with proof-based findings and a privacy gateway that keeps your data local.

Project last updated:08/29/26

GitHub Stars

878

Forks

153

Contributors

16

License

GPL-3.0

Why we included this project

Most pentesting setups hand you a pile of scanners and leave the orchestration to a human. DarkMoon flips that: point it at an authorized target and the self-hosted platform handles the assessment from start to finish, with roughly 50 specialist agents chaining real exploits across web apps, cloud, Active Directory, Kubernetes and other surfaces, then reporting every finding with proof rather than a vague summary. The privacy angle is what stands out for strict perimeters. The model only sees placeholder IPs, hosts and credentials, which are rehydrated locally, so sensitive data never leaves your network, and you can run the reasoning on a local or private LLM. That autonomy, plus a controlled MCP layer for executing real offensive operations, makes it worth evaluating alongside conventional scanners if you own authorized penetration testing or defensive validation.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category