Skills for Real Engineers. Straight from my .agents directory.
ponytail-review
Agent skill fromDietrichGebert/ponytail
Quality review of a change: is the logic right, is it safe, does it hold under real load, is risky code tested, is it fast enough, and is every line needed. Reads the connected code, not only the diff. Each finding is explained in plain English. Use for "review this", "code review", "review the last commit", "review my PR", "is this over-engineered", /ponytail-review.
How to install ponytail-review
npx skills add DietrichGebert/ponytail# or add only this skill to Claude Code: copy the folder
skills/ponytail-review/
# into
~/.claude/skills/ponytail-review/ (all projects)
.claude/skills/ponytail-review/ (one project)
1Review a change like the senior developer who will be paged when it breaks.2Order of importance: correct, safe, holds under load, tested, fast, lean.3Lean still matters: every extra line must be read, tested and fixed later.4This is a report the user asked for, so give it in full.56## 1. Understand first78- Review what the user names: uncommitted or staged changes, a branch, a PR9 link, or files. Nothing named: the uncommitted changes, or the last commit10 if there are none.11- Read the diff, then the code it touches: callers of every changed function,12 the functions it calls, the tests, the README.13- Trace the real flow: where data comes in, what is stored, what goes out.14- A change can break code it does not touch. When a signature, return value15 or behavior changes, grep every caller.16- Find the expected load in the repo (README, deploy config): one person17 running a script, or many users and processes at once. Judge scale against18 that, and say which load you assumed.1920## 2. Look for21221. **Bug:** wrong result, crash, missed edge case (empty, zero, last item,23 rounding, time zones), a caller broken by the change, a fix applied in one24 caller while the shared function stays broken.252. **Risk:** security holes (injection, weak randomness, secrets, missing26 checks on input from users), data loss (errors swallowed, writes in the27 wrong order, no transaction).283. **Scale:** fine for one user, wrong for many: check-then-write races, the29…
Excerpt shown under the repository's MIT license.
More skills in ponytail
Similar skill repositories
An agentic skills framework & software development methodology that works.
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
Public repository for Agent Skills
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Production-grade engineering skills for AI coding agents.
