GH Repository · 0xSteph
pentest-ai
Open-source AI pentester that proves every finding. Machine oracles re-run each exploit; verified bugs ship a proof capsule you can replay yourself.
- stars
- 1,677
- 30-day movement
- +103/day
- Related entries
- 60
- Connections
- 2
mcp-serverdevsecopsethical-hackingappsecai-pentestingai-penetration-testingdockersecuritypython/uvai-securitypythonvulnerability-scannerbug-bountycybersecurityctfPythonpenetration-testingclaudehacking-toolsred-teammcpmodel-context-protocoloffensive-securitypentesting
pentest-ai is an open-source AI-powered penetration testing tool that runs as an MCP server, usable with Claude. Each finding is verified by machine oracles that re-run the exploit, and verified bugs ship a replayable proof capsule.
Reach for it when you want AI-assisted pentesting where every reported bug comes with independently re-runnable proof rather than a bare claim.
Use it to
- Verify AI-reported vulnerabilities via replayable proof capsules
- Run AI-driven pentests through an MCP server
- Practice CTF-style offensive security tasks
- Assist bug bounty triage with exploit verification
- Integrate pentesting tools into Claude workflows
For Security researchers, pentesters, and bug bounty hunters
- Role
- mcp-server
- Language
- Python
- Licence
- MIT
- Forks
- 314
- Open issues
- 1
- Last push
- 2026-09-13
- Latest release
- v0.10.2 · 2026-05-10
topicsai-pentestingoffensive-securitymcp-serverpenetration-testingbug-bountyvulnerability-scanner