{"slug":"yogsoth-ai-validity-probing","source_name":"yogsoth-ai/validity-probing","name":"Yogsoth AI/Validity Probing","description":"Challenge construct validity — does benchmark measure claimed capability? — 3 benchmarks, 40 papers, 30 web searches","version":1,"lift":{"pass_rate_delta_pts":-60,"pass_rate_pct":58.3,"total_cases":24,"passed_cases":14,"tokens_delta_pct":98.5,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-09-04T05:19:13.015642+00:00"},"skill_score":null,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":-60,"with_pass_pct":20,"without_pass_pct":80,"tokens_delta_pct":98.5,"turns_delta_pct":0,"total_cases":24,"cases_aggregated":18,"verdict":"mixed","never_hurt":false,"completed_at":"2026-09-04T05:19:13.015642+00:00","run_id":"53b78654-b869-4744-a12a-b200b4222bcb","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":"Apache-2.0","install_count":0,"manifest_hash":"fd9c1f624158fcfd7e80d6ff6f7425e40bf4189b941e508b06035e329115e2ae","raw_url":"https://app.decimal.ai/s/yogsoth-ai-validity-probing/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/yogsoth-ai-validity-probing"}