{"slug":"agent-benchmark-suite","source_name":"agent-benchmark-suite","name":"Agent Benchmark Suite","description":"Agent skill for benchmark-suite - invoke with $agent-benchmark-suite","version":1,"lift":{"pass_rate_delta_pts":54.55,"pass_rate_pct":63.6,"total_cases":22,"passed_cases":14,"tokens_delta_pct":177.6,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-07-28T14:07:11.035976+00:00"},"skill_score":0.6364,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":54.55,"with_pass_pct":63.6,"without_pass_pct":9.1,"tokens_delta_pct":177.6,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":20,"verdict":"mixed","never_hurt":true,"completed_at":"2026-07-28T14:07:11.035976+00:00","run_id":"784d480a-cf31-4ddd-b82e-d22720670ce3","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":"589d37d00dcf1ed046e62ba647b90c1cdc50a78cb4777804fb75d2fedaf2fbab","raw_url":"https://app.decimal.ai/s/agent-benchmark-suite/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/agent-benchmark-suite"}