{"slug":"tencent-skillhone-evaluation","source_name":"tencent/skillhone-evaluation","name":"Tencent/Skillhone Evaluation","description":"Run and interpret skill evaluations. Use when you need to evaluate a skill, run probe/test/PR-val, check if a PR regresses quality, compare two versions, or diagnose why the score dropped. Handles the full eval lifecycle including solver trajectory diagnosis for tool-level error detection.","version":1,"lift":{"pass_rate_delta_pts":45.45,"pass_rate_pct":86.4,"total_cases":22,"passed_cases":19,"tokens_delta_pct":44.7,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-03T20:50:01.568085+00:00"},"skill_score":0.8636,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":45.45,"with_pass_pct":86.4,"without_pass_pct":40.9,"tokens_delta_pct":44.7,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":18,"verdict":"mixed","never_hurt":false,"completed_at":"2026-08-03T20:50:01.568085+00:00","run_id":"586afa90-d14d-4758-af1f-3743735a1ed3","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":"NOASSERTION","install_count":0,"manifest_hash":"1df75cbc20c053c71e5c7dc52ee3bec9e12c4898619d5de2cfd33633d68a9d27","raw_url":"https://app.decimal.ai/s/tencent-skillhone-evaluation/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/tencent-skillhone-evaluation"}