| name | ponytail-audit |
| description | Whole-repo audit for over-engineering — ranked delete/simplify list (repo-wide ponytail-review). Use for "audit this codebase", "find bloat", "ponytail-audit", "/ponytail-audit". Read-only; does not apply fixes. |
On-demand loading: Read when the user wants a repo-wide complexity audit, not a diff review.
When to use
- "audit this codebase", "audit for over-engineering", "what can I delete", "find bloat", "ponytail-audit", "/ponytail-audit"
- Before a large refactor to identify deletion candidates
Activation: Over-engineering and complexity only. Correctness, security, and performance are out of scope — route to code-reviewer, security-auditor, or differential-review.
Adapted from ponytail-audit (MIT).
Tags (same as ponytail-review)
delete: — dead code, unused flexibility, speculative feature
stdlib: — hand-rolled thing the stdlib ships
native: — dependency duplicating platform features
yagni: — abstraction with one implementation
shrink: — same logic, fewer lines
Process
- Scan the repo tree (exclude
node_modules, .git, dist, build, .cursor/plans, lockfiles).
- Hunt: unused deps, single-implementation interfaces, delegate-only wrappers, dead flags, hand-rolled stdlib.
- One line per finding, ranked biggest cut first:
path: tag: finding. replacement.
- End with
net: -N lines, -M deps possible. or Lean already. Ship.
Boundaries
Read-only — lists findings, applies nothing unless user asks to fix.