{"slug":"mkurman-benchmark-contamination-scan","source_name":"mkurman/benchmark-contamination-scan","name":"Mkurman/Benchmark Contamination Scan","description":"Scan training data for benchmark contamination — n-gram overlap, embedding similarity, and exact canary detection against 60+ evaluation datasets.","version":1,"lift":{"pass_rate_delta_pts":40.91,"pass_rate_pct":90.9,"total_cases":22,"passed_cases":20,"tokens_delta_pct":25.1,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-21T07:32:50.648320+00:00"},"skill_score":0.9091,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":40.91,"with_pass_pct":90.9,"without_pass_pct":50,"tokens_delta_pct":25.1,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":22,"verdict":"mixed","never_hurt":false,"completed_at":"2026-08-21T07:32:50.648320+00:00","run_id":"53e26fa0-dd26-4b36-8eba-9600c86ca465","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":"c4204567fe78709b7f3206235f97eb2bfd10b6204fa39094c8ae37409247998a","raw_url":"https://app.decimal.ai/s/mkurman-benchmark-contamination-scan/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/mkurman-benchmark-contamination-scan"}