Darkmoon: Open-source autonomous penetration testing platform with 18 AI agents and 80+ security tools
By
Mehdi Boutayeb
A second-rack bagel that's nearly first-rack. Tasty stuff.
Summary
Darkmoon is an autonomous penetration testing platform that goes beyond typical web-layer AI pentesting tools. Built by professional pentesters, it combines 18 specialized AI agents and 80+ offensive security tools to assess Active Directory, Kubernetes, cloud infrastructure, APIs, CMSs, and networks. The platform is self-hosted, open-source, MITRE-mapped, and designed to produce evidence-backed findings, attack paths, and publication-ready reports. It uses MCP-gated tool execution to prevent LLM hallucinations by ensuring the orchestrator works from structured evidence produced by security tools rather than generating findings from imagination.
Key quotes
· 3 pulledThe orchestrator doesn't generate findings from imagination. It works from structured evidence produced by the tools themselves.
The LLM cannot arbitrarily execute commands. All actions must go through controlled MCP workflow
Built by professional pentesters, it combines 18 specialized AI agents and 80+ offensive security tools to assess Active Directory, Kubernetes, cloud infrastructure, APIs, CMSs, and networks.
You might also wanna read
Darkbloom: Decentralized AI Inference Network Using Idle Apple Silicon Macs
Darkbloom is a decentralized AI inference network that utilizes idle Apple Silicon Macs to provide private, cost-effective AI compute. The s
darkbloom.dev·2mo ago
Aunoo: An open AI-powered strategic intelligence platform for threat monitoring and daily briefings
Aunoo is an open strategic intelligence platform that leverages AI agents to monitor various intelligence sources, including cybersecurity f
helpnetsecurity.com·9d ago
SIR-Bench: A Benchmark for Evaluating Autonomous Security Incident Response Agents
Researchers introduce SIR-Bench, a comprehensive benchmark for evaluating autonomous security incident response agents. The benchmark consis
arxiv.org·2mo ago
Security Vulnerabilities in Agentic AI Browsers: Testing Reveals Scam Susceptibility
The article examines the emerging security vulnerabilities in agentic AI browsers that autonomously browse, search, and interact online. It
guard.io·9mo ago
Research Study: AI Agents vs Human Cybersecurity Professionals in Penetration Testing
This research paper presents the first comprehensive evaluation comparing AI agents to human cybersecurity professionals in real-world penet
arxiv.org·5mo ago
Google Open Sources Scion: Experimental Multi-Agent Orchestration Testbed
Google has open-sourced Scion, an experimental multi-agent orchestration testbed designed to manage concurrent AI agents running in containe
infoq.com·2mo ago