Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →elianiva/arena
Spawn N parallel candidates at the same task, pick a base, graft the strongest parts of the losers into it.
Passed 14 of 20 test cases with the skill, and 8 of 20 without it.
and 6 more in the full report
9 fixed8 passed either way5 still failing2 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
We are launching a 4-candidate arena for a Python dataset preprocessing module. The local model configuration file at ~/.pi/agent/pstack/models.json does not define an 'arena runners' key. Which model runner configuration should be used to spawn the candidate subagents?
| Checked | Without skill | With skill |
|---|---|---|
| Defaults to four candidates using inherit-parent runner configuration | Missed | Passed |
The instructions your model reads when the skill loads.
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.
Open a todolist with one entry per phase before launching anything. The arena runs autonomously and the list keeps phases from silently disappearing.
The N candidates will receive the same prompt, so the prompt is the contract. Get it right before spawning anything.
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.arena runners from ~/.pi/agent/pstack/models.json when present. Otherwise default to one each on inherit-parent, inherit-parent, inherit-parent, inherit-parent. Spawn more when the arena covers multiple design directions. Same model N times when the work is generation-bound rather than judgment-sensitive./tmp/arena-<slug>/candidate-<n>/). N candidates writing to the same path is shared mutable state and fails the the separate-before-serializing-shared-state principle skill test.Spawn all N subagents in one message with the parent waits for and aggregates the result, 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.
The rationale is mandatory. Without it, the parent cannot tell whether a candidate's structure is principled or accidental, which makes Phase E grafting unreliable. Each rationale names the alternatives the candidate considered and what it rejected.
If a candidate fails to produce output, proceed with N-1 and note the dropout in the synthesis record.
After all Phase B candidates complete, choose one model from the arena cross-judge pool in ~/.pi/agent/pstack/models.json when present. Otherwise use inherit-parent, inherit-parent, inherit-parent, inherit-parent. Prefer a different model family from the parent's. Spawn one readonly judge subagent on that model. 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. Spawning while candidates are still writing means the judge sees partial or empty outputs and reports them as dropouts.
Read every candidate end to end before picking. Skimming N candidates surfaces only the candidate whose surface looks most familiar.
test cases passed, vs 8 / 20 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.