{"slug":"clawbio-clawpathy-autoresearch","source_name":"clawbio/clawpathy-autoresearch","name":"Clawbio/Clawpathy Autoresearch","description":"Eval-driven skill tuning. Given a task and an LLM-judge rubric, iteratively rewrites a SKILL.md until a downstream executor agent performs well against the judge. Low-code: all evaluation is LLM-as-judge, not deterministic Python.","version":1,"lift":{"pass_rate_delta_pts":25,"pass_rate_pct":79.2,"total_cases":24,"passed_cases":19,"tokens_delta_pct":9.8,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-16T06:23:22.344304+00:00"},"skill_score":0.7917,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":25,"with_pass_pct":79.2,"without_pass_pct":54.2,"tokens_delta_pct":9.8,"turns_delta_pct":0,"total_cases":24,"cases_aggregated":21,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-16T06:23:22.344304+00:00","run_id":"1d94db17-bd21-48e7-95a4-eb6e2444aaa5","version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":"MIT","install_count":0,"manifest_hash":"5bcab26daa156a862ac1ed5fff13fab58422240cfa3c3ca22b5eab79836d3d6a","raw_url":"https://app.decimal.ai/s/clawbio-clawpathy-autoresearch/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/clawbio-clawpathy-autoresearch"}