scaryrawr Swarm
Fan out N parallel Task workers. 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.
- Frame
- Fan out
- Aggregate
- Report
Phase A: Frame
- State the done predicate and the artifact or report the swarm must return.
- 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, orbest-ofbefore spawning. - Set N from the user or derive it from the shape. N is total workers, not the Task concurrency limit.
- Pick the worker model from
swarm workersininstructions/pstack-models.instructions.mdin Copilot home ($COPILOT_HOME, or$HOME/.copilotwhen unset) when present. Otherwise omitmodel. For a model race, name each arm's model up front. - Give every writing worker an isolated workspace: a distinct worktree and branch for repository changes, or a worker-specific directory under the session artifact directory for scratch output. Separate filenames inside one checkout are not isolation because workers would still share the working tree and Git index.
Phase B: Fan out
Spawn all N workers in one message with agent_type: "general-purpose" and mode: "background". Pass the configured model unless it is absent or set to auto.
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.
If a worker drops out, proceed with N-1 and note it.
Phase C: Aggregate
Read the terminal results. 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.