arena · diff

git:20260923.11fd787 to git:20261014.8882c71

14 added, 37 removed. Audit A to A.

---

name: arena

description: "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."

disable-model-invocation: true

---

# Arena

- Fan out N parallel attempts at the same task. Read every candidate end to end. Pick the strongest as the base. Graft the best ideas from the others into it. Verify the synthesized result.

- ## Start

- Open a todolist with one entry per phase before launching anything.

- 1. Frame

- 2. Fan out

- 3. Cross-judge

- 4. Pick

- 5. Graft

- 6. Verify

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

## Phase A: Frame

- The N candidates will receive the same prompt, so the prompt is the contract.

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

1. State the artifact each candidate is producing.

- 2. Derive the rubric. State what success looks like for *this* task, then turn it into 3-6 concrete gradeable criteria. The rubric is the picker's tool in Phase D. Candidates only see the task.

- 3. Pick the runners from the `arena runners` role in `${CLAUDE_SKILL_DIR}/../poteto-mode/references/models.md`, one per entry. Spawn more when the arena covers multiple design directions. Same model N times when the work is generation-bound rather than judgment-sensitive.

- 4. Assign output paths. Each candidate writes to its own location (`isolation: "worktree"` where possible, otherwise `/tmp/arena-<slug>/candidate-<n>/`), per **principle-separate-before-serializing-shared-state**. Principles live at `${CLAUDE_SKILL_DIR}/../poteto-mode/principles/<name>.md`.

+ 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` role in `${CLAUDE_SKILL_DIR}/../poteto-mode/references/models.md`, one per entry. Spawn more runners when the arena covers multiple design directions. Use the same model N times when the work is generation-bound rather than judgment-sensitive.

+ 4. Give each candidate its own output location (`isolation: "worktree"` where possible, otherwise `/tmp/arena-<slug>/candidate-<n>/`), per **principle-separate-before-serializing-shared-state**. Principles live at `${CLAUDE_SKILL_DIR}/../poteto-mode/principles/<name>.md`.

## Phase B: Fan out

- Runners load no pstack skill, so make every skill-relative path and principle name in the task absolute before you spawn them. Spawn all N subagents with the Agent tool 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 both the artifact and a short rationale.

- Each rationale names the alternatives the candidate considered and what it rejected.

+ Runners load no pstack skill, so make every skill-relative path and principle name in the task absolute before you spawn them. Spawn all N subagents with the Agent tool 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 fails to produce output, proceed with N-1 and note the dropout in the synthesis record.

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

## Phase C: Cross-judge

- After all Phase B candidates complete, choose one model from the `arena cross-judge pool` role in `models.md`, one that differs from the parent's model when possible. Spawn one judge on that model with `subagent_type: "pstack:read-only"`. It sees the rubric and the candidates by path label, scores each criterion, and recommends a base with rationale. It runs in parallel with the parent's reading in Phase D, not with the candidates themselves. Don't spawn the judge while candidates are still writing.

+ After all Phase B candidates complete, choose one model from the `arena cross-judge pool` role in `models.md`, one that differs from the parent's model when possible. Spawn one judge on that model with `subagent_type: "pstack:read-only"`. It sees the rubric and the candidates by path label, scores each criterion, and recommends a base with rationale. It runs in parallel with your reading in Phase D. Don't spawn it while candidates are still writing.

## Phase D: Pick a base

- Read every candidate end to end before picking.

- Score each candidate against the rubric criterion by criterion, not on holistic feel. Compare against the cross-judge. 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 on which candidate a future maintainer can extend most easily without breaking invariants. Prefer the cleaner boundary or smaller API when two feel tied, per the Laziness Protocol.

+ 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. Agreement on the base confirms the pick. Disagreement means one of you is biased or the rubric was ambiguous. Read both rationales before deciding.

- Record the pick and the reason in a short synthesis note alongside the base artifact, including the cross-judge's verdict.

+ 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 and identify what is worth porting into the base. The signal is usually one or two things per candidate, not most of it.

- Fold each graft in by hand, per **principle-redesign-from-first-principles**. Don't paste mechanically. The result has to remain coherent under one mental model.

- Record what was grafted, from which candidate, and what was rejected and why.

+ 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 **principle-redesign-from-first-principles**, so the result stays coherent under one mental model. Don't paste mechanically.

- When N candidates converge on the same shape, that is a strong agreement signal. Note the convergence in the record and ship the consensus shape. No graft is needed. When N candidates wildly diverge, Phase A was under-specified. Reframe and re-run rather than averaging the divergence.

+ 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

- The synthesized artifact has to hold up under the same scrutiny as any other output, per **principle-prove-it-works**.

- If verification surfaces a problem the arena did not catch, either Phase A was wrong (re-frame and re-run) or one candidate caught it and you missed the graft (go back to Phase E). Don't paper over.

+ Verify the synthesized artifact like any other output, per **principle-prove-it-works**. 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. One short synthesis note alongside, naming the base, the grafts (with source candidate), the rejections, the dropouts if any, and the verification result.

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