{"skill":{"slug":"guard","displayName":"Guard","summary":"Deep AI safety guardrails workflow—policy definition, input/output filtering, monitoring, escalation, and false-positive handling. Use when reducing harmful...","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":242,"installsAllTime":1,"installsCurrent":1,"stars":0,"versions":1},"createdAt":1774632828879,"updatedAt":1774633009145},"latestVersion":{"version":"1.0.0","createdAt":1774632828879,"changelog":"Version 1.0.0 – Initial Release\n\n- Introduces a comprehensive deep AI safety guardrails workflow for LLM-based products.\n- Details a six-stage process: policy scope, threat modeling, controls stack, implementation patterns, monitoring & review, and iteration & appeals.\n- Provides specific guidance on policy definition, input/output filtering, monitoring, escalation, and false-positive handling.\n- Includes review checklist and tips for best practices in deploying safety guardrails for AI features.\n- Addresses enterprise-specific considerations (e.g., data-leak prevention for internal bots).","license":"MIT-0"},"metadata":null,"owner":{"handle":"clawkk","userId":"s170g5yz1q3ksjnn4gz6v24af983h1mh","displayName":"clawkk","image":"https://avatars.githubusercontent.com/u/265748372?v=4"},"moderation":null}