Swarm

Claude Code: read Cursor terms in this skill through claude-code/runtime.md (tools, subagents, models, transcripts).

Fan out N parallel workers, each in its own git worktree. They may cover separate slices, race the same brief, or mix both. The parent waits, aggregates, and returns one report.

Start

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

1. Frame
2. Fan out
3. Aggregate
4. Report

Phase A: Frame

1. State the done predicate and the artifact or report the swarm must return.
2. Choose the shape. Partition into slices, race N workers on identical briefs, or mix both. For a race or mixed shape, declare first pass, rank all, or best-of before spawning.
3. Set N from the user or derive it from the shape. N is total workers, not the concurrent subagent limit.
4. Resolve the swarm workers model/effort pair from the live delegation-kit table when present, otherwise from ~/.claude/rules/pstack-models.md; default model: sonnet, effort: medium. Map auto, inherit-parent, and inherit to that explicit pair. Fan-outs of 3+ use sonnet/haiku seats, never opus. Pass both fields on every call and retry. Name each race arm's pair up front.
5. Give each worker its own writable output when it writes. When workers verify or measure commits, each brief names the exact SHAs. A measurement brief also names the method (sample count, what one sample is, order). The worker records both in its result.

Phase B: Fan out

Spawn all N workers in one message with subagent_type: general-purpose, isolation: "worktree", run_in_background: true, and explicit model plus effort from step 4. Omit isolation only when the worker must share the parent's checkout.

When a worker must start from a non-default pushed branch, tell it to fetch and check out that branch inside its worktree.

Every brief stands alone. Include the goal, scope, exact slice or race arm, how to verify, and what to report. Reports use PASS, ISSUES, or BLOCKED with evidence. A worker that can prove a defect reports ISSUES and lists every issue it can prove, not only the first.

If a worker drops out, proceed with N-1 and note it.

Phase C: Aggregate

Read the terminal results. Drop a result that does not record the SHAs and method its brief names, and respawn that worker once. After a second miss, record a gap. A gap does not count as a pass. For coverage, every required slice needs a result. For a race, apply the selection rule declared up front. Use first pass, rank all, or best-of. Do not paste raw worker dumps.

Keep a compact result table, one-line evidenced issues, and explicit gaps or dropouts.

Phase D: Report

Return one consolidated in-chat report with the table, issue one-liners, gaps or dropouts, and the race rule when used.