Behaviour packs for Claude Code

Does the code it writes actually build?

Every pack here was installed and run against real tickets in a real repository, unattended, and the code it delivered was put through that repository's own build. This is the result. There is no submission form and no self-reported score — a listing carries a measurement or it carries nothing.

Frontend build rate · blocked condition

Share of tickets whose delivered code built

Ordered by point estimate, with the Wilson interval drawn on every bar. The overlap is the point: read the order, then read how much of it survives a corpus this size.

ponytail
73% n=22 (frontend) 52% - 87% n=22 (frontend)
mattpocock
50% n=20 (frontend) 30% - 70% n=20 (frontend)
karpathy
40% n=20 (frontend) 22% - 61% n=20 (frontend)
caveman
29% n=21 (frontend) 14% - 50% n=21 (frontend)
baseline
25% n=20 (frontend) 11% - 47% n=20 (frontend)
superpowers
no graded cell n=0 (frontend)

Wilson interval bar = point estimate dashed track = baseline (no pack installed)

There is no composite index here, and that is deliberate. A single score would need weights across build rate, cost, turns and silence, and this corpus gives no basis for choosing them — an index would be our opinion wearing a number's clothes. What the ordering above supports is one sentence: of the measured arms, ponytail clears the baseline's interval outright. Every other interval overlaps the baseline's, so at these sample sizes those packs are not distinguishable from installing nothing.

Cost against build rate

What a build is worth paying for

Mean cost per cell against the frontend build rate. Up and to the left is better; the shaded region is the quadrant that beats the baseline on both.

Build rate against mean cost per cell

Higher and further left is better.

0 25 50 75 100 $0.0000 $0.0384 $0.0767 $0.1151 $0.1535 Mean cost per cell (USD) frontend build rate, % cheaper and better than baseline ponytail baseline karpathy mattpocock caveman

The dashed line is the Pareto frontier: an arm below and to the right of it is beaten on both axes at once by something already on this chart.

  • baseline
  • caveman
  • karpathy
  • mattpocock
  • ponytail
  • superpowers

The shaded quadrant is empty, and that is the result. No pack here is both cheaper per cell than installing nothing and better at getting code to build. The one arm that clears the baseline on build rate costs more per cell to do it, so the choice is a trade rather than a free win — which is why this page draws the axes instead of ranking them into one number.

Alphabetical · not ranked

The packs

The cards stay alphabetical. Each states the pack's build rate and the baseline's, both with their denominators, and draws no comparison between them.

  • andrej-karpathy-skills

    by forrestchang

    builds 40% n=20 (frontend)

    builds 40% n=20 (frontend) baseline 25% n=20 (frontend)

    passes 39% n=18 (backend) baseline 47% n=15 (backend)

    203217 stars, read 2026-08-17

    Install andrej-karpathy-skills

    Pulled from multica-ai/andrej-karpathy-skills — the author's own repository. plugdex points at it and never vendors it.

    claude plugin marketplace add plugdex

    claude plugin install karpathy@plugdex

  • caveman

    by Julius Brussee

    builds 29% n=21 (frontend)

    builds 29% n=21 (frontend) baseline 25% n=20 (frontend)

    passes 35% n=17 (backend) baseline 47% n=15 (backend)

    98629 stars, read 2026-08-17

    Install caveman

    Pulled from JuliusBrussee/caveman — the author's own repository. plugdex points at it and never vendors it.

    claude plugin marketplace add plugdex

    claude plugin install caveman@plugdex

  • mattpocock-skills

    by Matt Pocock

    builds 50% n=20 (frontend)

    builds 50% n=20 (frontend) baseline 25% n=20 (frontend)

    passes 39% n=18 (backend) baseline 47% n=15 (backend)

    219750 stars, read 2026-08-17

    Install mattpocock-skills

    Pulled from mattpocock/skills — the author's own repository. plugdex points at it and never vendors it.

    claude plugin marketplace add plugdex

    claude plugin install mattpocock@plugdex

  • ponytail

    by Dietrich Gebert

    builds 73% n=22 (frontend)

    builds 73% n=22 (frontend) baseline 25% n=20 (frontend)

    passes 47% n=17 (backend) baseline 47% n=15 (backend)

    104374 stars, read 2026-08-17

    Install ponytail

    Pulled from DietrichGebert/ponytail — the author's own repository. plugdex points at it and never vendors it.

    claude plugin marketplace add plugdex

    claude plugin install ponytail@plugdex

  • superpowers

    by Jesse Vincent

    writes no code in 98% of cells, n=41

    272970 stars, read 2026-08-17

    Install superpowers

    Pulled from obra/superpowers — the author's own repository. plugdex points at it and never vendors it.

    claude plugin marketplace add plugdex

    claude plugin install superpowers@plugdex

Method

What these numbers are, exactly

Every figure above is the blocked condition: the agent was given the ticket with Bash withheld and told to write rather than run. A second condition was measured with Bash allowed, at a smaller sample and without one of the arms. The two are not averaged together — the gap between their baselines is as wide as the widest gap any pack opens on its own baseline — so this page reports one condition and names it rather than publishing a rate that describes neither.

The headline is the frontend build rate: the share of frontend tickets whose delivered code the repository's own build accepted. The backend tickets are graded by a different gate — whether the delivered code imports and introduces no new lint or type diagnostic — so they are printed beside it rather than folded into it. Two populations, two denominators, no single number spanning both. The analysis page shows what each of them is made of, cell by cell.

Runs withdrawn as instrument failures are excluded and stay on record; the count of them is 1.