{"slug":"mohitagw15856-ai-eval-plan","source_name":"mohitagw15856/ai-eval-plan","name":"Mohitagw15856/AI Eval Plan","description":"Design an evaluation plan for an LLM or AI feature before shipping it. Use when asked how to evaluate a prompt/model/agent, set up an eval harness, define quality metrics for an AI feature, or build a regression gate. Produces an eval plan — task definition, datasets, metrics & rubrics, baselines, automated + human evals, a pass bar, and a regression gate.","version":1,"lift":{"pass_rate_delta_pts":45.45,"pass_rate_pct":63.6,"total_cases":22,"passed_cases":14,"tokens_delta_pct":14.3,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-14T01:57:06.035363+00:00"},"skill_score":0.6364,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":45.45,"with_pass_pct":63.6,"without_pass_pct":18.2,"tokens_delta_pct":14.3,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":22,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-14T01:57:06.035363+00:00","run_id":"662fce37-51b1-4976-8367-e41fecc08c8b","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":"56ddeaed79d3dcc6c7ebc9df2a35d1f701ece2834b68cdc497d2cd2d2885855d","raw_url":"https://app.decimal.ai/s/mohitagw15856-ai-eval-plan/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/mohitagw15856-ai-eval-plan"}