{"skill":{"slug":"model-tester","displayName":"Model Tester","summary":"Test agents or models against predefined test cases to validate model routing, performance, and output quality. Use when: (1) verifying a specific agent or m...","tags":{"agents":"1.0.0","debugging":"1.0.0","latest":"1.0.0","testing":"1.0.0"},"stats":{"comments":0,"downloads":387,"installsAllTime":3,"installsCurrent":3,"stars":0,"versions":1},"createdAt":1773232981985,"updatedAt":1777528335911},"latestVersion":{"version":"1.0.0","createdAt":1773232981985,"changelog":"Initial release of model-tester.\n\n- Provides a command-line tool to validate agents or models against predefined test cases.\n- Supports testing model routing, performance, and output quality with structured JSON reporting.\n- Allows targeting specific agents or models using `--agent`, `--model`, and `--case` parameters.\n- Extracts actual model usage, token counts, and runtime from OpenClaw logs for verification.\n- Ensures privacy by using only static prompts and structured log fields—no user data involved.","license":"MIT-0"},"metadata":null,"owner":{"handle":"nandorocker","userId":"publishers:nandorocker","displayName":"Nando Rossi","image":"https://avatars.githubusercontent.com/u/1476836?v=4"},"moderation":{"isSuspicious":true,"isMalwareBlocked":false,"verdict":"suspicious","reasonCodes":["suspicious.llm_suspicious"],"summary":"Detected: suspicious.llm_suspicious","engineVersion":"v2.4.5","updatedAt":1777528335911}}