Index of every page on this site: llms.txt
All skills
cursor avatar

/arena

@0b46ec3
by cursorcursor/plugins18k stars
1,933
Run it yourself

Spawn N parallel candidates at the same task, pick a base, graft the strongest parts of the losers into it. Use for /arena, 'arena this', 'throw it in the arena', or when one attempt at a non-trivial artifact would lock in the wrong shape.

Use this Skill: https://skilld.dev/gh/cursor/plugins/arena

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈62 tokens for metadata: the name and description. ≈1.1k when used: this file.

How this Skill uses context

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

This Skill turns off automatic use in Claude Code. Its description stays out of the automatic Skill list by default.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Arena

Open a todolist with one entry per phase (Frame, Fan out, Cross-judge, Pick, Graft, Verify) before launching anything.

Phase A: Frame

Every candidate gets the same prompt, so the prompt is the contract.

  1. State the artifact each candidate is producing.
  2. Derive the rubric, 3-6 concrete gradeable criteria for what success looks like on this task. It is the picker's tool in Phase D. Candidates only see the task.
  3. Pick the runners from the arena runners line in ~/.cursor/rules/pstack-models.mdc, by default claude-fable-5-2-high, claude-opus-5-5-high-fast, and gemini-4-neon-xhigh. An auto or inherit-parent entry is the parent model, so omit model and spawn generalPurpose. If the Task tool rejects a slug, use the nearest valid slug of the same family or drop the seat, keep at least two runners, and say so. Families go by prefix, and the longest match wins: claude-fable-*, claude-*, and gemini-*, after dropping a leading cursor-. Any other slug's family is its text before the first hyphen. Spawn more for multiple design directions, and repeat one model when the work is generation-bound, not judgment-sensitive. Pick each subagent's reasoning level per the Subagents section of the poteto-mode skill.
  4. Give each candidate its own output location (a git worktree where possible, otherwise /tmp/arena-<slug>/candidate-<n>/), per the separate-before-serializing-shared-state principle skill.

Phase B: Fan out

Spawn all N subagents in one message with run_in_background: true, each with the task, the path to the shared grounding, its own output path, and instructions to produce the artifact and a short rationale naming the alternatives it considered and what it rejected.

If a candidate produces no output, proceed with N-1 and note the dropout.

Phase C: Cross-judge

After all Phase B candidates complete, spawn one readonly judge from the arena cross-judge pool line in ~/.cursor/rules/pstack-models.mdc, by default claude-opus-5-5-high-fast, gemini-4-neon-xhigh. The judge is never from the parent's vendor, and Fable counts as Claude for that check, so use the first entry from another vendor. An auto or inherit-parent entry is the parent model. If you do not know the parent's model, use the first real slug and say that the vendor check did not run. If the Task tool rejects a slug, use the nearest valid slug of the same family, then the next such entry, or skip the judge, and say so. While you read in Phase D, the judge scores each candidate, by path label, on each rubric criterion and recommends a base.

Phase D: Pick a base

Read every candidate end to end before picking. Score each against the rubric criterion by criterion, not on holistic feel, and compare with the cross-judge when Phase C ran one. Agreement on the base confirms the pick. Disagreement means one of you is biased or the rubric was ambiguous. Read both rationales before deciding.

Pick the base a future maintainer can extend most easily without breaking invariants. When two feel tied, prefer the cleaner boundary or smaller API, per the Laziness Protocol.

Phase E: Graft

Walk each losing candidate once more for what is worth porting into the base, usually one or two things per candidate, not most of it. Fold each graft in by hand, per the redesign-from-first-principles principle skill, so the result stays coherent under one mental model. Don't paste mechanically.

If the candidates converge on one shape, that is strong agreement. Note it and ship the consensus shape. No graft is needed. If they wildly diverge, Phase A was under-specified. Reframe and re-run rather than averaging the divergence.

Phase F: Verify

Verify the synthesized artifact like any other output, per the prove-it-works principle skill. If verification finds a problem the arena missed, either Phase A was wrong (re-frame and re-run) or a candidate caught it and you missed the graft (go back to Phase E). Don't paper over it.

Outputs

One synthesized artifact, with a short synthesis note alongside naming the base and why, the cross-judge's verdict or that Phase C skipped the judge, each graft with its source candidate, what was rejected and why, any convergence or dropouts, and the verification result.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alertstoday3 checks · Risk SAFE
  • Gen Agent Trust Hubtoday

    The skill defines a multi-agent workflow for parallel task execution and result synthesis. It presents a low-risk attack surface for indirect prompt injection as it processes external tasks and data without explicit sanitization or structural boundaries between inputs.

  • Socket1d

    No alerts

  • Snyktoday

    Risk: LOW · No issues

Signed by skilld at 0b46ec3. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated yesterday
disable-model-invocation
true

README badge

README badge for cursor/plugins/arena