ponytail-review

Reviews code diffs for over-engineering and lists what to delete or simplify.

Updated Aug 7, 2026
One-click install
npx skills add https://github.com/Sambhav242005/Major-Project --skill ponytail-review-sambhav242005
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ponytail-review
Source: https://github.com/Sambhav242005/Major-Project/tree/main/.agents/skills/ponytail-review
Command: npx skills add https://github.com/Sambhav242005/Major-Project --skill ponytail-review-sambhav242005

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Codebases accumulate unnecessary complexity: hand-rolled utilities that duplicate the standard library, speculative abstractions with a single implementation, and dependencies that add nothing. This Skill reviews diffs exclusively for over-engineering and tells you exactly what to cut, keeping reviews focused and actionable. ## Core Features & Use Cases - Complexity-Only Review: Hunts dead code, reinvented stdlib functionality, unneeded dependencies, and speculative abstractions while explicitly excluding correctness, security, and performance concerns. - One-Line Findings: Each finding uses a compact tagged format (delete, stdlib, native, yagni, shrink) with location, what to cut, and what replaces it. - Net-Line Scoring: Ends every review with a single metric estimating how many lines the diff could lose. - Use Case: Before merging a pull request that adds a custom retry wrapper, an abstract repository with one implementation, and a moment.js import, run this review to get findings like "L4: native: moment.js imported for one format call. Intl.DateTimeFormat, 0 deps." ## Quick Start Ask the AI to review this diff for over-engineering and tell me what can be deleted.

Frequently Asked Questions about ponytail-review

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

FAQPage Schema
How do I review code for over-engineering?

Invoke the review on a diff and it returns one line per finding: location, a tag (delete, stdlib, native, yagni, shrink), what to cut, and what replaces it. It ends with a net line-reduction estimate so you can judge the simplification potential.

What kinds of complexity does an over-engineering review find?

It finds hand-rolled code the standard library already ships, dependencies duplicating native platform features, abstractions with a single implementation, dead code, and logic that can be expressed in fewer lines.

Does this review catch bugs or security issues?

No. Correctness bugs, security holes, and performance problems are explicitly out of scope and should be routed to a normal review pass. This review only hunts unnecessary complexity.

Will the review apply the suggested deletions automatically?

No. It only lists findings with locations and replacements; it never modifies the code. You decide which suggestions to apply in a separate editing step.

When should I not use an over-engineering review?

Skip it when a diff needs correctness, security, or performance validation, since those are out of scope. It also never flags minimal smoke tests or assert-based self-checks, which it treats as the acceptable minimum.