Detects prompt injection by its effect on a sacrificial canary model, not just pattern matching: untrusted input hits a powerless model first, a behavioral check reads the residue, and it returns block, flag, or pass before your primary model acts. Inbound preflight sensor, not a guarantee.

CVE hunting harness for Claude Code - 20 skills, 5-agent team, systematic vulnerability research with false positive elimination

Lightweight Agent Detection & Response (ADR) layer for AI agents — guards commands, files, and web requests. Part of Gen Agent Trust Hub.