{"slug":"ruvnet-harness-bench","source_name":"ruvnet/harness-bench","name":"Ruvnet/Harness Bench","description":"Manage `@metaharness/darwin` bench suites — `bench create <repo>` scaffolds a JSON suite from a repo's test corpus; `bench verify <suite.json>` checks suite well-formedness. Bench suites are the fixed evaluation corpora that `harness-evolve --bench <suite.json>` scores variants against, decoupling evolution from the repo's natural tests. Degrades gracefully when @metaharness/darwin is absent.","version":1,"lift":{"pass_rate_delta_pts":45.45,"pass_rate_pct":86.4,"total_cases":22,"passed_cases":19,"tokens_delta_pct":-37.4,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-07T21:14:18.300877+00:00"},"skill_score":0.8636,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":45.45,"with_pass_pct":86.4,"without_pass_pct":40.9,"tokens_delta_pct":-37.4,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":19,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-07T21:14:18.300877+00:00","run_id":"e030bb49-d09a-4228-a7cf-07cb676e4d6b","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":"5248e16b24c8742ae2bbf636a1839179c1e7ed07af5e807a8eeccf8bfcf7f176","raw_url":"https://app.decimal.ai/s/ruvnet-harness-bench/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/ruvnet-harness-bench"}