{"slug":"borghei-agent-harness","source_name":"borghei/agent-harness","name":"Borghei/Agent Harness","description":"Test and evaluation harness for AI agents — scenario suites, deterministic replay, regression diffing, cost and latency budgets. Use when agent quality is vibe-checked, before shipping a prompt or model change, or when evals drift.","version":1,"lift":{"pass_rate_delta_pts":31.82,"pass_rate_pct":77.3,"total_cases":22,"passed_cases":17,"tokens_delta_pct":98.6,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-20T06:38:45.197785+00:00"},"skill_score":null,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":31.82,"with_pass_pct":77.3,"without_pass_pct":45.5,"tokens_delta_pct":98.6,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":21,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-20T06:38:45.197785+00:00","run_id":"6955c00f-bc7c-4373-ac93-116186f5feff","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 + Commons Clause","install_count":0,"manifest_hash":"b44573fc4a14680d617dcb8d418eb9306751c61f778b11452e94fbe375edb70e","raw_url":"https://app.decimal.ai/s/borghei-agent-harness/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/borghei-agent-harness"}