Browse Claude Skill projects under the "agent-security-eval" topic.
TrustedExecBench: Scenario-grounded security evaluation for autonomous personal AI assistants.