🛡️ A firewall for your agent's brain. Open-source security proxy that protects AI agents from prompt injection attacks. 4-stage pipeline, community threat feed, no bypass mode.
Stars
15
Forks
1
Watchers
15
Open Issues
0
Overall repository health assessment
^2.3.432
commits
docs: update CHANGELOG with false positive fix and community threat feed verification
8ecda76View on GitHubfeat: add /api/scan endpoint for arbitrary text scanning (stages 2-4)
cdea905View on GitHub