memtrapEvaluate and harden AI agent memory against DeepMind traps and OWASP ASI06 attacks, scoring resistance and providing automated protections.
agentcopCatches prompt injection hidden in untrusted external content, tool results, and agent-to-agent communication — plus credential exfiltration, token smuggling...
agentverifSCAN → SIGN → VERIFY. Certify your skill, detect tampering, revoke instantly. Full control over how your skill is distributed and run. Requires AGENTVERIF_AP...