Audits an entire codebase for over-engineering and ranks findings to delete or simplify.
npx skills add https://github.com/lrstanley/skills --skill ponytail-audit-lrstanley
Please help me install this Agent Skill. Skill: ponytail-audit Source: https://github.com/lrstanley/skills/tree/main/ponytail-audit Command: npx skills add https://github.com/lrstanley/skills --skill ponytail-audit-lrstanley
delete, stdlib, native, yagni, or shrink, with a concrete replacement named. - Ranked One-Shot Report: Outputs one line per finding ordered biggest cut first, ending with a net line/dependency savings estimate. - Use Case: Before a major refactor, run the audit to identify wrappers that only delegate, factories with one product, and dependencies the platform already provides, then hand the list to your team for cleanup. ## Quick Start Ask the AI to run ponytail-audit on this repository and list what can be deleted or replaced with standard library equivalents.High-intent search queries and answers about installing and using this skill.
Run ponytail-audit against the repository root. It scans the whole tree and returns a ranked list of findings tagged delete, stdlib, native, yagni, or shrink, each with a concrete replacement suggestion and file path.
The audit hunts for dead flags, unused config, wrappers that only delegate, and dependencies duplicating platform or standard library features. Findings are ranked biggest cut first with an estimated net line and dependency reduction.
ponytail-review analyzes a single diff, while ponytail-audit scans the entire codebase. Both use the same tagging system, but the audit produces a one-shot whole-repo report instead of reviewing incremental changes.
No. It only lists findings and applies nothing. The report is one-shot, and you decide which recommendations to implement manually or in a separate change pass.
Correctness bugs, security vulnerabilities, and performance problems are explicitly out of scope. Those should be routed to a normal code review pass; the audit covers only complexity and bloat.
arbiterForgeIsolate untrusted repositories in ephemeral Docker sandboxes for inspection and execution.
oporpinoVerifies tests and integrates completed work via PR, merge, or discard.
Jatinkrmahato993203Reviews code diffs for over-engineering and lists what to delete or simplify.
cursorRemoves unjustified code comments by auditing a diff with a review subagent and fixing root causes.