# Swarm

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

- Skill: `shishiv/swarm` (Agent Skill)
- Install (CLI): `npx skillmds@latest add shishiv/swarm`
- Raw SKILL.md: https://api.skillmd.com/api/skills/shishiv/swarm/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Docs & Writing
- Author: shishiv (https://skillmd.com/u/shishiv)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/shishiv/swarm

---


# Swarm

Fan out N parallel managed-worktree 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. Use the implementation role by default. For a model race, inspect the models reported by the host, name each available model up front, and pass it explicitly on that arm. Never invent a model ID.
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

Follow [`../../docs/delegation.md`](../../docs/delegation.md). Launch the
workers together with stable task keys. Give every writer its own worktree.
Use a local delegate when the work needs access to the user's computer.

When a worker must start from a non-default pushed branch, pass the managed
worktree base branch explicitly.

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.