{"slug":"eval-driven-dev","source_name":"eval-driven-dev","name":"Eval Driven Dev","description":"You're building an **automated evaluation pipeline** that tests a Python-based AI application end-to-end — running it the same way a real user would, with real inputs — then scoring the outputs using evaluators and producing pass/fail results via `pixie test`.","version":1,"lift":{"pass_rate_delta_pts":33.33,"pass_rate_pct":33.3,"total_cases":24,"passed_cases":8,"tokens_delta_pct":132,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-07-23T23:41:13.727148+00:00"},"skill_score":0.3333,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":33.33,"with_pass_pct":33.3,"without_pass_pct":0,"tokens_delta_pct":132,"turns_delta_pct":0,"total_cases":24,"cases_aggregated":19,"verdict":"mixed","never_hurt":false,"completed_at":"2026-07-23T23:41:13.727148+00:00","run_id":"759e683a-ac51-4538-8a25-c4448de16b00","version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":null,"content_status":"clean","indexable":true},"license":"MIT","install_count":0,"manifest_hash":"8ae2e579275f3d6a6b570fe9642771aad83fb7bafbef6f2bbb9dd12d69b56a3e","raw_url":"https://app.decimal.ai/s/eval-driven-dev/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/eval-driven-dev"}