{"slug":"kunanonj-agent-gan-evaluator","source_name":"kunanonj/agent-gan-evaluator","name":"Kunanonj/Agent Gan Evaluator","description":"GAN Harness — Evaluator agent. Tests the live running application via Playwright, scores against rubric, and provides actionable feedback to the Generator.","version":1,"lift":{"pass_rate_delta_pts":36.36,"pass_rate_pct":86.4,"total_cases":22,"passed_cases":19,"tokens_delta_pct":52.2,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-07T00:03:31.451897+00:00"},"skill_score":0.8636,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":36.36,"with_pass_pct":86.4,"without_pass_pct":50,"tokens_delta_pct":52.2,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":19,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-07T00:03:31.451897+00:00","run_id":"03f13085-ac1c-41f1-97cf-475e9108beb3","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","install_count":0,"manifest_hash":"9573ac446d06abc7ccc698bd04767e3b08ecd5468697a2d556e5eba0c957ddcf","raw_url":"https://app.decimal.ai/s/kunanonj-agent-gan-evaluator/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/kunanonj-agent-gan-evaluator"}