{"slug":"affaan-m-eval-harness","source_name":"affaan-m/eval-harness","name":"Affaan M/Eval Harness","description":"Formal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles","version":15,"lift":{"pass_rate_delta_pts":22.73,"pass_rate_pct":95.5,"total_cases":22,"passed_cases":21,"tokens_delta_pct":46,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-03T18:13:22.009466+00:00"},"skill_score":0.9545,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":22.73,"with_pass_pct":95.5,"without_pass_pct":72.7,"tokens_delta_pct":46,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":22,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-03T18:13:22.009466+00:00","run_id":"6abb001d-e06c-4193-b439-0706508a8fec","version_number":15,"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":"a60ca764c99d93d54d9419f8436a1e3f9ab0eb645c2e2477e9fde09ff8b5157f","raw_url":"https://app.decimal.ai/s/affaan-m-eval-harness/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/affaan-m-eval-harness"}