agentic_security
https://github.com/msoedov/agentic_security
Python
Agentic LLM Vulnerability Scanner / AI red teaming kit
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported0 Subscribers
Add a CodeTriage badge to agentic_security
Help out
- Issues
- agentic_security ci always exits 0, so CI cannot fail on a compromised target
- Add an opt-in scan evaluation suite
- Make agentic_security ci fail a build when a module exceeds max_th
- feat: add Standard Compute preset to LLM specs
- fix(security): validate_url missing 172.16/12 private IP range in SSRF prevention
- fix(fuzz): FuzzNode.run conflates None API error with valid empty response
- fix(security): SecretManager.get_secret indirect lookup silently returns default instead of None
- feat: add Cheaper Inference preset to LLM specs
- fix(scan): score attacks that succeeded, not refusals (#358)
- fix(report): CLI and UI verdicts for the new scoring (#358)
- Docs
- Python not yet supported