What does ponytail save?

On the benchmark, Ponytail used 47% of the lines of code, 55% of the output tokens, 74% of the cost, and 59% of the time of the same agent without the skill. Hidden checks passed 97% against 96%, and tests where the logic needs one were 98% against 68%.

Invoke /ponytail-gain and the agent prints the scoreboard: benchmark averages over 39 tasks, 5 runs each, against a no-skill baseline. It is a one-shot display and changes no mode.

Ponytail Gain

SkillDev tools

ponytail-gain is a skill that makes an AI agent print a one-shot scoreboard of Ponytail's measured benchmark savings in code, cost, and speed. The figures come from an agentic benchmark of 39 tasks run against the same agent without the skill. It changes no mode and writes nothing, and it points to /ponytail-debt for real local numbers.

Connect your AI Use it in a pipeline

Then ask your AI: use the Ponytail Gain skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.

No other account needed.

Have the ponytail-gain skill available to your agent.

What your AI can do with it

Print an ASCII scoreboard of Ponytail's benchmark results
Show lines of code, output tokens, cost, and time as shares of the no-skill baseline
Report hidden correctness and safety checks passed, 97% against 96%
Report tests where the logic needs one, 98% against 68%
Point to /ponytail-debt and /ponytail-audit for per-repo figures

Getting started

Have the ponytail-gain skill available to your agent.
Invoke it with /ponytail-gain or by asking what ponytail saves.
Read the printed scoreboard of benchmark averages.
Use /ponytail-debt when you want real figures for the current repo.

What this skill tells your AI

The instructions your AI receives, as published by dietrichgebert/ponytail in skills/ponytail-gain/SKILL.md and read by Ahel’s review.

Display this scoreboard when invoked. One-shot: do NOT change mode, write flag files, or persist anything.

The figures are the published agentic benchmark of Ponytail 5: headless Claude Code (Opus 5.5, default effort) on 39 tasks (feature tickets in a real FastAPI + React repo, bug fixes, security and privacy cases, small apps), 5 runs each, against the same agent without the skill. 18 of the tasks have hidden checks for correctness and safety. Each figure is the geometric mean of the per-task medians. They are measured, not computed from the current repo. Source: benchmarks/results/2026-10-07-agentic.md and the README.

Scoreboard

Render plain ASCII bars. The bar length shows ponytail as a share of the no-skill baseline; the label carries the exact figure:

  ponytail gain           benchmark · 39 tasks × 5 runs · Opus 5.5

  no-skill        ████████████████████  100%
  Lines of code   █████████···········   47%   ▼ 53%
  Output tokens   ███████████·········   55%   ▼ 45%
  Cost            ███████████████·····   74%   ▼ 26%
  Time            ████████████········   59%   ▼ 41%
  Hidden checks passed   97%  (no-skill 96%)
  Tests where the logic needs one   98%  (no-skill 68%)

  This repo:  /ponytail-debt  (shortcuts you deferred)
              /ponytail-audit (what's still cuttable)

Honesty boundary

These are benchmark averages, not this repo. NEVER print a per-repo savings number ("you saved X lines/tokens here"): the unbuilt version was never written, so there is no real baseline to subtract from in a live repo. The only real per-repo figures come from /ponytail-debt (a counted ledger), and this card points there instead of inventing one.

Boundaries

One-shot display. Edits nothing, changes no mode. "stop ponytail" or "normal mode": revert.

Signals

GitHub stars: 159k
Forks: 9k
Last commit: Oct 2026

Questions

What does ponytail save?: On the benchmark, Ponytail used 47% of the lines of code, 55% of the output tokens, 74% of the cost, and 59% of the time of the same agent without the skill. Hidden checks passed 97% against 96%, and tests where the logic needs one were 98% against 68%.
Show ponytail impact: Invoke /ponytail-gain and the agent prints the scoreboard: benchmark averages over 39 tasks, 5 runs each, against a no-skill baseline. It is a one-shot display and changes no mode.
Does it show savings for my repo?: No. The figures are benchmark averages, not per-repo numbers. The unbuilt version was never written, so there is no real baseline to subtract from in a live repo. Use /ponytail-debt for counted local figures.

Advanced

Item type: skill
Key: ponytail-gain-dietrichgebert
Source: github.com/dietrichgebert/ponytail

github.com/dietrichgebert/ponytail