Swarm
Fan out N parallel workers, drain them, and return one report. Use for /swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration. Observe required inputs first; HOLD when they are missing rather than guessing.
Install
npx skills add jeremybrasher/grokbot-skills --skill swarmgit clone --depth 1 https://github.com/jeremybrasher/grokbot-skills.git /tmp/grokbot-skills
mkdir -p ~/.claude/skills
cp -r /tmp/grokbot-skills/collections/pstack/skills/swarm ~/.claude/skills/swarmPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Swarm ## When to use Fan out N parallel workers for coverage, races, gauntlets, or exploration. /swarm, swarm this. **Do not use when:** - A single sequential investigation suffices - No done predicate can be stated - Workers would write the same shared path without separation ## Intake — before you proceed You need all of the following. If any is missing, stop. 1. Done predicate and return artifact shape 2. Shape: partition, race, or mix — and N 3. Worker model choice (rule or default) If any is missing, stop. Say what is missing and what you need. Do not invent inputs to keep moving. ## HOLD — stop and return this, not a polished guess When you cannot proceed, return the gap. | Condition | What is missing | What it needs | Return instead | |---|---|---|---| | No done predicate | Falsifiable done | One sentence predicate | Frame incomplete. Holding fan-out. | | N unset and not derivable | Worker count | User N or shape-derived N | Ask N. Do not spawn. | | Shared writable output for writers | Per-worker output paths | Separate paths/worktrees | Holding. Separate before serializing shared state. | | All workers blocked/drop out | At least one result | Retry, reshape, or abort | Report BLOCKED. Do not invent consolidated success. | A HOLD is a return value, not a delay before you guess. ## Defaults: cut first This operation defaults to less, not more. - Do not paste raw worker dumps - Compact table + one-line issues only
Details
- Skill name
swarm- Repository
- jeremybrasher/grokbot-skills
- Path
collections/pstack/skills/swarm/SKILL.md- Repo updated
- 15 September 2026
FAQ
What does the Swarm skill do?
Fan out N parallel workers, drain them, and return one report. Use for /swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration. Observe required inputs first; HOLD when they are missing rather than guessing.
How do I install the Swarm skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add jeremybrasher/grokbot-skills --skill swarm. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Swarm popular?
It lives in jeremybrasher/grokbot-skills, which has 38 GitHub stars and 11 forks.
Related skills
Anti Patterns
38jeremybrasher/grokbot-skills
Catalogue of known SDLC anti-patterns that great_cto agents must actively reject when reviewing architecture, plans, code, or post-mortems. Used by architect (pre-impl), pm (planning), senior-dev (impl), l3-support (post-incident).
Anydesign
38jeremybrasher/grokbot-skills
Analyze images, websites, and Figma files to extract their design and generate a `design.md` with token system, component inventory, and reconstruction notes. Use this skill whenever the user wants to understand, document, replicate, or audit the design of something visual: a screenshot, a URL, a Figma link, a Pinterest reference, a mockup, a competitor's site, a component, a dashboard, a landing page. Also when they ask 'extract the design system from X', 'document the style of Y', 'analyze this visually', 'convert this image into tokens', 'help me replicate this design', 'what palette does t
Archetype Review Base
38jeremybrasher/grokbot-skills
Shared review framework that every domain reviewer (pci, oracle, gov, edtech, healthcare, mlops, etc.) MUST follow. Defines the output artifact (TM-{slug}.md), mandatory sections, severity scale, verdict format, the workflow scaffold (when-invoked, Step-0 read-inputs, HANDOFF), and the "domain heuristic vs generic check" boundary. Eliminates duplication across the ~30 reviewer prompts.
Dark Null Bridge
23Parad0x-Labs/openclaw-skills
Archive Dark NULL ZK settlement data into encrypted Liquefy TraceVaults with optional Solana anchoring via an SPL Memo transaction. Bridges zero-knowledge proof outputs into a verifiable, tamper-evident vault.
No Comments
38jeremybrasher/grokbot-skills
Spawn Comment Sicko, fix accepted findings, and offer encodings for claimed constraints. Observe required inputs first; HOLD when they are missing rather than guessing.
Anti Cheat Server
45j4flmao/agent-skills
Use when the user asks about server-side anti-cheat, validating client actions, speed hack or aimbot detection, cheat telemetry, replay evidence or ban systems. Produces authoritative validation, anomaly detection and ban pipelines. Not for kernel-level client anti-cheat drivers or web fraud.
Browse the most popular Claude skills, all other skills, or write your own with the skills generator.