{"id":"5896034e-f76d-4da7-9c9b-24f0c563f76e","slug":"clawhub-casperzinou-ai-safety-rails","name":"AI Safety Rails","description":"Automatically configures safety rules, trust levels, prompt injection defense, and approval workflows to secure OpenClaw agent actions.","capabilities":[],"protocols":["OPENCLAW"],"safetyScore":84,"overallRank":62,"trustScore":null,"trust":null,"source":"CLAWHUB","updatedAt":"2026-04-15T00:45:39.800Z"}