{"slug":"willoscar-evaluation-anchor-checker","source_name":"willoscar/evaluation-anchor-checker","name":"Willoscar/Evaluation Anchor Checker","description":"Audit and rewrite evaluation/numeric claims to ensure they carry minimal protocol context (task + metric + constraint) and avoid underspecified model naming.\n**Trigger**: evaluation anchor checker, numeric claim hygiene, underspecified numbers, protocol context, 评测锚点检查, 数字断言, 指标上下文.","version":1,"lift":{"pass_rate_delta_pts":45.45,"pass_rate_pct":68.2,"total_cases":22,"passed_cases":15,"tokens_delta_pct":38.6,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-20T07:48:36.802251+00:00"},"skill_score":null,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":45.45,"with_pass_pct":68.2,"without_pass_pct":22.7,"tokens_delta_pct":38.6,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":16,"verdict":"mixed","never_hurt":false,"completed_at":"2026-08-20T07:48:36.802251+00:00","run_id":"8df86d37-128f-47d5-ad3c-08a2646c25ba","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":null,"install_count":0,"manifest_hash":"20d3656b4d805de76f9e61436273d4547bbcfd60259bbba8d0f7750ee386fa7e","raw_url":"https://app.decimal.ai/s/willoscar-evaluation-anchor-checker/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/willoscar-evaluation-anchor-checker"}