ponytail-audit

Run a whole-repo audit that scans every file for dead code, single-implementation abstractions, and hand-rolled utilities duplicating the standard library. The output is a ranked list of findings tagged by type, with a concrete replacement and file path for each.

Look for factories with one product, interfaces with a single implementation, wrappers that only delegate, dead config flags, and files exporting one thing. The audit tags these as `delete` or `yagni` findings and ranks the biggest cuts first.

ponytail-review scans a diff, while ponytail-audit scans the entire repository tree. Both use the same tagging system (delete, stdlib, native, yagni, shrink), but the audit produces a one-shot whole-repo report instead of reviewing changes.

No. The audit only lists findings and applies nothing. It is a one-shot report describing what to cut and what to replace it with, leaving all changes to the developer.

Correctness bugs, security vulnerabilities, and performance problems are explicitly out of scope. The audit covers only over-engineering and complexity; other issues should go through a normal review pass.