{"slug":"ruvnet-cost-benchmark","source_name":"ruvnet/cost-benchmark","name":"Ruvnet/Cost Benchmark","description":"Run the corpus benchmark — booster locally, optional Gemini/Sonnet/Opus baselines — and persist a verifiable measured-vs-claimed table","version":1,"lift":{"pass_rate_delta_pts":77.27,"pass_rate_pct":90.9,"total_cases":22,"passed_cases":20,"tokens_delta_pct":-18.9,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-07T20:49:23.412320+00:00"},"skill_score":0.9091,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":77.27,"with_pass_pct":90.9,"without_pass_pct":13.6,"tokens_delta_pct":-18.9,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":20,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-07T20:49:23.412320+00:00","run_id":"f3d98f79-9db4-4f32-89e1-d92e8d80f325","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":"cf56dc2d590e3109023f86d70b6bd707b3072b339df7c304149acaaa6cb52820","raw_url":"https://app.decimal.ai/s/ruvnet-cost-benchmark/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/ruvnet-cost-benchmark"}