arenalisted

# 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. The arena runs autonomously and the list keeps phases from silently disappearing. 1. Frame 2. Fan out 3. Cross-judge 4. Pick 5. Graft 6. Verify ## Phase A: Frame The N candidates will receive the same prompt, so the prompt is the contract. Get it right before spawning anything. 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. Concrete: `Adds a --dry-run flag that skips writes`. Vague: `code is correct`. The rubric is the picker's tool in Phase D; candidates only see the task. 3. Pick the runners. Spawn each candidate with the Agent tool using `subagent_type: general-purpose` and a `model` override. Default to one each on `opus`, `fable`, and `sonnet`. For genuine cross-family diversity, run one candidate through the OpenAI Codex CLI (`pm-skill:codex-help` shows the invocation) and treat its output as a fourth candidate. 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 (a git worktree where possible, otherwise `/tmp/arena-