arena

Files (1)

• SKILL.md

SKILL.md (first lines)

# Arena

Other pstack skills named here are files, not Skill tool calls, and the Skill tool refuses them. To run or apply one, Read `~/.claude/skills/<name>/SKILL.md` in full, or `~/.claude/skills/principle-<name>/SKILL.md` for a principle.

Open a task list (`TaskCreate` and `TaskUpdate`, or `TodoWrite` where that is the todo tool) 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. Use the `arena runners` line in `~/.claude/pstack/models.md`. If the file or that line is missing, default to one each on `fable high` and `opus xhigh`. Pass each entry's model and effort as the Agent call's `model` and `effort`. An `inherit` entry in this line or the cross-judge line means the parent session's model, so omit `model` and `effort` for it. If an entry names a model other than `fable`, `opus`, or `inherit`, run that seat on `fable high` and say so. 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. Pick each subagent's effort for the job, 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. A worktree starts from the repo's default branch, not the current one, so a worktree candidate's brief names your HEAD SHA (`git rev-parse HEAD`) and tells it to run `git checkout --detach <sha>` in its worktree first.

## 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, choose one entry from the `arena cross-judge pool` line in `~/.claude/pstack/models.md`. If the file or that line is missing, choose from `fable high` and `opus xhigh`. Prefer a model that differs from the parent's. Spawn one judge subagent with `subagent_type: "pstack-readonly"` on that entry's `model` and `effort`. 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 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.

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, each graft with its source candidate, what was rejected and why, any convergence or dropouts, and the verification result.

From RayenTellissy/pstack-claude, MIT.

More from RayenTellissy/pstack-claude