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.
Something's coming. Be the first to know.→the_whole_idea
- class CacheManager:- def __init__(self, ttl, maxsize):- self._store, self._lock = {}, Lock()- # ...44 more lines you maintain forever+ @lru_cache(maxsize=1000)+ def fetch(...): ...
// same behavior. 48 fewer lines. zero bugs in code that no longer exists.
the_ladder
<input type="date"> over a picker lib.benchmarks
// medians across 12 feature tasks on a FastAPI + React repo.
// validation, error handling, security and accessibility are never simplified away.
install
codex plugin marketplace add DietrichGebert/ponytailcopilot plugin install ponytail@ponytailgemini extensions install github.com/DietrichGebert/ponytailpi install git:github.com/DietrichGebert/ponytail// also OpenCode, Cursor, Windsurf, Cline, Kiro, Zed and more — 14+ agents.
// full list in the README.
commands
/ponytail lite|full|ultra|off | set intensity, or turn it off |
/ponytail-review | find over-engineering in the current diff |
/ponytail-audit | scan the whole repo for bloat |
/ponytail-debt | collect deferred shortcuts into a ledger |
/ponytail-gain | show the benchmark scoreboard |
/ponytail-help | quick command reference |
intensity
Builds what you asked, names the lazier alternative in one line. You pick.
The ladder, enforced. Stdlib and native first. Shortest diff, shortest explanation.
YAGNI extremist. Ships the one-liner and challenges the rest of the requirement in the same breath.