メニュー
Audit an answer or implementation against primary sources before recommending changes.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
Evaluate benchmark design for fair baselines, measurable outcomes, leakage, cherry-picking, and reproducibility.
Review a large question by staying inside one assigned slice, then support synthesis across several focused shard answers.
Check whether a change is ready for a public release candidate across packaging, docs, tests, and rollback risk.
Review code, docs, or plans for correctness risks, missing evidence, unclear tradeoffs, and untested assumptions.
SOC 職業分類に基づく
| name | source-grounded-audit |
| description | Audit an answer or implementation against primary sources before recommending changes. |
| license | MIT |
Use this when the answer depends on an external standard, product behavior, public API, registry convention, or repository contract.
Before recommending a design:
When reviewing another model's answer, start with the strongest source mismatch. Then give the smallest correction that makes the answer fit the documented behavior.
