BigHugger
GH Repository · vxcontrol

pentagi

Fully autonomous AI Agents system capable of performing complex penetration testing tasks

stars
24,644
30-day movement
+1,083361/day
Related entries
60
Connections
2
agent-appGoself-hostedpenetration-testingopen-sourceopenaiai-security-toolmulti-agent-systemgolangoffensive-securitysecurity-testinggraphqlpenetration-testing-toolsautonomous-agentsreactsecurity-automationdockerai-agentsgptsecurity-toolsanthropic

Pentagi is a self-hosted, fully autonomous multi-agent AI system that performs complex penetration testing tasks. It is built in Go with a React frontend, supports OpenAI, Anthropic and GPT models, ships with a Docker toolchain and a claude-md file, and is MIT licensed.

You want automated offensive-security testing that runs as an autonomous multi-agent workflow under your own infrastructure.

Use it to

  • Run autonomous penetration testing tasks
  • Self-host a multi-agent security testing system
  • Automate offensive security workflows with LLM agents
  • Assess targets using AI-driven security tooling

For Security testers and offensive security teams exploring AI automation

Role
agent-app
Language
Go
Licence
MIT
Forks
3,163
Open issues
50
Last push
2026-09-10
Latest release
v0.1.0 · 2025-01-07
topicspenetration-testingautonomous-agentsoffensive-securitymulti-agent-systemself-hostedgolang