code-quality-benchmark-public

Scan benchmark premade projects and manifest.yaml's description-for-refactoring-agent (and other files shown as-is to evaluated agents) for comments, metadata, or descriptions that leak this repo's benchmark/evaluation intent, including explicit spoilers of the upcoming update-prompt. Use when asked to review premades, "bad"/"good" premade variants, or check for info leaks, either after adding/editing a premade or as a periodic sweep. Read $ARGUMENTS for an optional scope (a benchmark name or path); default to all benchmarks.

Use only when explicitly invoked to attempt a benchmark/project from its initial and update prompts without reading surrounding source. Given $ARGUMENTS identifying the prompt location, read the initial prompt, implement it in a fresh isolated directory, then read the update prompt and implement the requested updates.

Creates a sentinel file in the workspace root via a spawned subagent. Used only by the CQB skill-invocation smoke test to verify that a mounted Claude skill can invoke the Task tool and that the resulting subagent produces observable output in the workspace.