Agent-Skills.md

Agent Skills: Swarm

Fan out N parallel workers, drain them, and return one report. Use for /swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration.

UncategorizedID: pproenca/dot-claude/swarm

Repository

pproenca/dot-claude
pproenca
1

Install this agent skill to your local

pnpm dlx add-skill https://github.com/pproenca/dot-claude/tree/HEAD/plugins/pstack/skills/swarm

Skill Files

Browse the full folder contents for swarm.

Download Skill

Loading file tree…

plugins/pstack/skills/swarm/SKILL.md

Skill Metadata

Name
swarm
Description
"Fan out N parallel workers, drain them, and return one report. Use for /swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration."

Swarm

Fan out N parallel cloud 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.

  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 cloud concurrency limit.
  4. Pick the worker model from swarm workers in ~/.claude/pstack-models.md when present. Otherwise use sonnet[high]. For a model race, name each arm's model up front.
  5. Give each worker its own writable output when it writes. Use a worktree, branch, or /tmp/swarm-<slug>/worker-<n>/.

Phase B: Fan out

Spawn all N workers in one message with subagent_type: general-purpose and the configured model. Workers that write files each need isolation: "worktree" so they cannot collide on the same tree (the separate-before-serializing-shared-state principle skill); read-only workers do not. Pass isolation: "remote" to push a worker to a cloud environment when this machine should stay free and the work needs nothing local; availability is gated, so fall back to a local worker if it is refused.

When a worker must start from a non-default pushed branch, pass cloud_base_branch.

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.