Display a text scoreboard of published Ponytail 5 benchmark results, comparing code volume, generated text tokens, cost, time, and check results against an agent without the skill. The figures describe benchmark averages, not savings in the current project.
Show ponytail measured impact as a scoreboard: less code, less cost, more speed, from the agentic benchmark averages. One-shot display.
This skill displays published Ponytail 5 benchmark results as a scoreboard with text bars. It compares lines of code, output tokens, cost, and time against an agent without the skill. Tokens are units used to count the amount of generated text. The display also includes the hidden-check pass rate and the proportion of logic that has tests where needed.
Use it to review Ponytail's measurements quickly or compare correctness and testing figures alongside code volume and cost. The display includes the task count, number of runs, and model used, so you can see the conditions behind the comparison.
The figures are averages from a published benchmark using headless Claude Code with Opus 5.5, running each of 39 tasks five times. This skill does not measure the current project or calculate individual savings. It displays the results once, without changing files or operating modes or saving state. The license is MIT.
This overview was written by AI (OpenAI Codex) from the public SKILL.md. Check the original below for the exact behavior.
AI rated this skill on four points using only the contents of its SKILL.md. Stars and popularity are not taken into account.
The total counts each point as A = 25, B = 15, C = 5. Rated by AI (OpenAI Codex) as of Oct 20, 2026. Treat it as a guide and read the original before installing.
Review the instructions the agent will receive before you install.
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.
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)
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.
One-shot display. Edits nothing, changes no mode.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Guide coding tasks toward the smallest complete change by reusing existing code, platform features, and installed dependencies. Address affected callers and necessary checks, then explain skipped work, unchecked areas, and risks in a brief reply.
Guide coding tasks toward the smallest complete change by checking affected code, reusing existing features, and avoiding unnecessary additions. It also calls for relevant checks and concise explanations of what changed, what remains unchecked, and any risks.
Audit an entire code repository or a selected folder for bugs, security risks, load problems, missing tests, slow paths, and excess code. Get up to 20 prioritized findings explaining concrete problems, minimal fixes, and consequences.
Review a codebase for bugs, security risks, failures under expected load, missing tests, slow paths, and unnecessary code. Receive prioritized findings with concrete failure cases, small proposed fixes, and the consequences of leaving each issue unresolved.
Collect comments marking deferred improvements into a ledger grouped by file. Record what was simplified, its stated limit, and when to revisit it, while flagging entries that lack a trigger for future action.
Collect comments marking deliberate coding shortcuts into a file-by-file debt ledger. The report shows each shortcut’s stated limit and revisit trigger, flags missing triggers, and leaves the code unchanged.