{"slug":"loulanyue-benchmark-performance-baseline-regression-detection","source_name":"loulanyue/benchmark-performance-baseline-regression-detection","name":"Loulanyue/Benchmark Performance Baseline Regression Detection","description":"- Before and after a PR to measure performance impact","version":1,"lift":{"pass_rate_delta_pts":54.55,"pass_rate_pct":95.5,"total_cases":22,"passed_cases":21,"tokens_delta_pct":-15.2,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-21T21:07:09.142751+00:00"},"skill_score":0.9545,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":54.55,"with_pass_pct":95.5,"without_pass_pct":40.9,"tokens_delta_pct":-15.2,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":22,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-21T21:07:09.142751+00:00","run_id":"bba239df-1825-497b-9430-fcf34a7610e8","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":"62ab5b3be34d8a6591f107884a84091349d7fcdcd7cac8c78d2b7a55de1d895e","raw_url":"https://app.decimal.ai/s/loulanyue-benchmark-performance-baseline-regression-detection/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/loulanyue-benchmark-performance-baseline-regression-detection"}