ponytail-audit

Audits an entire codebase for over-engineering and ranks deletable code, dependencies, and abstractions.

Updated May 11, 2026
One-click install
npx skills add https://github.com/cloudofgeorge/AI-hands-Engineer --skill ponytail-audit-cloudofgeorge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ponytail-audit
Source: https://github.com/cloudofgeorge/AI-hands-Engineer/tree/main/skills/engineering/optimization/ponytail/ponytail-audit
Command: npx skills add https://github.com/cloudofgeorge/AI-hands-Engineer --skill ponytail-audit-cloudofgeorge

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Codebases accumulate dead code, redundant dependencies, and speculative abstractions over time, and finding this bloat manually across a whole repository is slow and inconsistent. This Skill scans the entire repo and produces a ranked list of what to delete, simplify, or replace with standard library or native platform equivalents. ## Core Features & Use Cases - Whole-repo over-engineering scan: Unlike a diff-based review, it audits the entire codebase tree in one pass. - Tagged findings: Each finding is labeled as delete (dead code), stdlib (hand-rolled standard library functionality), native (redundant platform features), yagni (unused abstractions), or shrink (shortenable logic). - Ranked one-shot report: Outputs one line per finding ordered by biggest cut first, ending with a net estimate of removable lines and dependencies. - Use Case: Before a major refactor, run the audit to identify single-implementation interfaces, one-product factories, dead config flags, and dependencies the standard library already covers, then hand the ranked list to your team for cleanup planning. ## Quick Start Ask the agent to run ponytail-audit on this repository to find over-engineering and list what can be deleted or replaced with stdlib equivalents.

Frequently Asked Questions about ponytail-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit a codebase for over-engineering?

Run a whole-repo audit that scans the entire tree for dead code, unused abstractions, and redundant dependencies. The audit tags each finding by category and ranks them biggest cut first, ending with an estimate of removable lines and dependencies.

What is the difference between ponytail-audit and ponytail-review?

ponytail-audit scans the entire codebase, while ponytail-review only examines a diff. Both use the same tagging system for dead code, stdlib replacements, native equivalents, yagni abstractions, and shrinkable logic.

What kinds of code bloat does a repo audit detect?

It detects dead code, dependencies duplicating standard library or platform features, single-implementation interfaces, one-product factories, pass-through wrappers, dead config flags, and hand-rolled stdlib functionality.

Does the over-engineering audit automatically fix or delete code?

No, it is a one-shot report that only lists findings and applies nothing. You review the ranked output and decide what to remove, and correctness bugs or security issues require a separate normal review pass.

When should I not use a whole-repo complexity audit?

Do not use it to find correctness bugs, security holes, or performance problems, since those belong in a normal review pass. It only targets complexity: things to delete, simplify, or replace with existing platform features.