Injects the ruleset every turn at the active level; adds the /ponytail commands (see Commands). OpenCode also auto-loads this repo's AGENTS.md, so the rules hold even without the plugin. The plugin adds the lite/full/ultra/off levels.

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. - ponytail/skills/ponytail at main · DietrichGebert/ponytail

ponytail He says nothing. He writes one line. It works. // A ruleset that makes your AI coding agent write the least code that works — like a senior dev who's been paged at 3am one too many times.

ponytail is the only arm that cuts every metric, and the only one that stays fully safe while doing it. The cut is biggest where there is a real over-build trap (date picker 404 to 23 lines, color picker 287 to 23, because it reaches for a native <input> instead of a component) and near zero on code that is already minimal.

With ponytail: ... More survivors in examples/. Numbers The honest measurement is a real agent doing real work: a headless Claude Code session editing tiangolo's full-stack-fastapi-template (a real FastAPI + React repo), scored on the git diff it leaves behind. Twelve feature tickets, the same agent with and without the skill, n=4 ...

AGENTS.md is a repository-level file—just commit it to git: ... All agents that read AGENTS.md (CodeWhale, Aider, Zed, Pi, Kiro, Codex extension) will automatically follow.

The Ponytail project injects a strict "lazy senior dev" ruleset into agents like Claude Code, Codex, and Gemini CLI to reduce generated code by 80–94%. The project is MIT-licensed and hosted on GitHub.

Ponytail Installation Guide: How to Use It with Codex, Claude Code, and Gemini CLI Ponytail is a rule and plugin set for AI coding agents. It helps Codex, Claude Code, Copilot CLI, Gemini CLI, and similar tools decide whether new code is really needed before writing it, and encourages reuse of existing code, standard libraries, platform features, and installed dependencies.

It includes commands for changing intensity, reviewing diffs, auditing repositories, and tracking deferred shortcuts. It is useful for developers who want AI coding tools to act more like pragmatic senior engineers and less like generators of unnecessary abstractions.

Fix it once, where all callers route through. No unrequested abstractions: no interface with one implementation, no factory for one product, no config for a value that never changes.