メニュー
Audit a RecurseKit run for correctness. Reviews events, skipped files, and result quality.
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。
SOC 職業分類に基づく
Run a Recursive Language Model loop over a large input. Use when content exceeds context window or when token-efficient processing of large codebases/documents is needed.
Run a Recursive Language Model loop over large inputs using the RecurseKit runtime.
| name | recurse-audit |
| description | Audit a RecurseKit run for correctness. Reviews events, skipped files, and result quality. |
| context | fork |
| allowed-tools | ["Read","Bash","Grep"] |
| argument-hint | Optionally specify a run-id to audit |
Audit a RecurseKit run:
$ARGUMENTS
python3 scripts/recurse-kit.py status --all to list runs..recurse-kit/workspace/<run-id>/progress.json -- did it complete?.recurse-kit/workspace/<run-id>/events.jsonl -- any errors or budget exceedances?.recurse-kit/workspace/<run-id>/skipped.log -- were relevant files excluded?.recurse-kit/workspace/<run-id>/report.json -- token usage and savings..recurse-kit/workspace/<run-id>/result.md -- is the answer complete and accurate?