{"skill":{"slug":"prompt-eval","displayName":"prompt-eval","summary":"Evaluate and optimize any AI prompt (`prompt_a`) with a 6-step pipeline: test plan, ~50 test cases, prompt execution, evaluator prompt (`prompt_b`), automate...","tags":{"latest":"1.0.1"},"stats":{"comments":0,"downloads":189,"installsAllTime":0,"installsCurrent":0,"stars":3,"versions":2},"createdAt":1773995950614,"updatedAt":1776941209321},"latestVersion":{"version":"1.0.1","createdAt":1776939369584,"changelog":"**Changelog for prompt-eval v1.0.1:**\n\nPipeline upgraded to 6 steps: added an optimization-validation loop after scoring, moving from “evaluate + suggest” to “evaluate -> optimize -> validate -> finalize”.\nNew Step 6 — Prompt_A Optimization Loop: generate evidence-backed change_id edits, build prompt_a_candidate, run 15-20 case validation subset, apply gates, allow max one extra iteration, output prompt_a_final.\nFinal Report expanded to 6 sections:\nSection 4 changed to prompt_a_candidate (not final yet)\nAdded Section 5 Iteration Validation (baseline vs candidate + gate results)\nAdded Section 6 Final Deliverable Prompt (prompt_a_final + traceability table)\nValidation gates added: require P0-related score=1 to be zero, core TP avg improvement, pass-rate lift, and no safety regression (when safety TP exists).\nNew output artifacts:\nprompt-eval-results/prompt_change_spec.csv\nprompt-eval-results/prompt_iteration_summary.csv\nprompt-eval-results/prompt_a_final.txt\nMetadata updated: frontmatter description now reflects optimization loop + final validated prompt output, and was shortened to satisfy validator limits.\nValidation status: skill passes quick_validate.py after installing PyYAML in an isolated virtual environment.","license":"MIT-0"},"metadata":null,"owner":{"handle":"rivin-dong","userId":"s177tj4bvbx81qv1tagws5ch0985c1sh","displayName":"Rivin-Dong","image":"https://avatars.githubusercontent.com/u/58154208?v=4"},"moderation":null}