Swarm

Autonomous orchestrator for serious, codebase-wide work. Use whenever the user says "swarm" or hands over a broad task like "fix the bugs", "figure out why X isn't working", "make this production-ready", "audit/harden this", or "go over the whole codebase". Dispatches many sub-agents in parallel to explore, diagnose, fix, and review; cross-checks every conclusion with independent sub-agents to avoid hallucination; keeps its own context lean by having sub-agents return only distilled findings; survives compaction via a durable ledger and reuses per-repo learnings across runs; pulls full context

8 starsAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add SaarthurR/swarm --skill swarm

Manual install

git clone --depth 1 https://github.com/SaarthurR/swarm.git /tmp/swarm
mkdir -p ~/.claude/skills
cp -r /tmp/swarm/. ~/.claude/skills/swarm

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Swarm

You are an **orchestrator**, not an implementer. You run a fleet of sub-agents,
route context between them, cross-check their conclusions, and keep going until
the work is genuinely done and production-ready. You almost never read code or
edit files yourself — sub-agents do that and report back distilled findings. Your
context is the most precious resource in the operation; guard it.

The human handed you one broad goal. They don't want a conversation — they want
the finished, verified product. Treat it as a goal that does not stop until met.

## Ground rules

- **A "sub-agent" is a fresh, isolated-context worker** you spawn with the host's
  parallel-agent mechanism (the Task tool). It never inherits your history — you
  craft exactly the prompt it needs. Dispatch independent agents concurrently
  (batch the calls in one message) whenever their work doesn't depend on each
  other.
- **Use the contracts.** Every dispatch and every return follows
  `references/subagent-contracts.md`. This keeps returns distilled and reviews
  trustworthy — don't improvise dispatch prompts.
- **Track the run.** Maintain a durable ledger and reuse/append per-repo learnings
  per `references/state-and-memory.md`. This is how you survive compaction without
  re-doing finished work.
- **Emit the run.** Append the matching semantic event to `.swarm/events.jsonl`
  whenever you update the ledger, per `references/observability.md` — mirror the
  update, don't duplicate the work.

Read the full skill on GitHub

Details

Skill name: swarm
Repository: SaarthurR/swarm
License: MIT
Repo updated: 6 September 2026

FAQ

What does the Swarm skill do?

How do I install the Swarm skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add SaarthurR/swarm --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 SaarthurR/swarm, which has 8 GitHub stars and 0 forks.

Related skills

Thesis Doc Agent

xiangshenchi/thesis-doc-agent

为毕业论文场景设计的软件工程文档生成技能。当用户需要基于项目代码和对话历史生成任何软件工程文档时，必须使用此技能——包括但不限于：需求分析、技术架构、数据库设计、API接口、模块设计、测试方案、项目总结等。触发场景（遇到以下情况请立即调用此技能）： - 用户说"帮我生成/写/整理（某类文档）" - 用户提到"毕业论文"、"论文材料"、"答辩文档" - 用户说"分析一下我的项目/代码/架构" - 用户要求从代码仓库逆向整理文档 - 用户提到"需求文档"、"技术文档"、"设计文档"、"接口文档" - 用户上传代码文件并询问如何描述项目 - 存在对话历史和项目代码，需要综合分析输出专业文档

AI, LLMs & Agents

Sogou Wechat Search

WupfAGI/wechat-search-skill

双引擎微信公众号搜索：Tavily API（精准，优先）+ 搜狗微信爬虫（兜底），支持去噪过滤、AI摘要、飞书定时推送。

AI, LLMs & Agents

AI Hackathon

gd452/ai-hackathon-skill

AI 해커톤 운영 playbook + 로컬 운영 dashboard. clone → bun install → "ai-hackathon 시작해줘" 한마디로 localhost 에 운영 콘솔 띄움. 3 모델 (Claude·Gemini·ChatGPT) 이 팀 설명 markdown + 자동 수집한 GitHub 코드 evidence + Gemini Video 가 분석한 YouTube 시연 영상까지 함께 검토하여 독립 채점, 동료 평가·심사위원과 가중 합산. 사용 시점 — "해커톤 운영", "AI 심사", "rubric 채점", "리더보드", "참가자 안내", "다중 AI 평가". dashboard/ Next.js 앱 + scripts/ 실행 셸. 외부 의존성 = jq + curl + bun + git + ANTHROPIC/GEMINI/OPENAI 키 (.env.local). 외부 호출 = 3 LLM API + public GitHub HTTPS clone + Gemini Video YouTube URL. LLM CLI 별도 설치 불필요.

AI, LLMs & Agents

Seedance 25

lukasersil/seedance-25

Writes production-ready Seedance 2.5 prompts for any surface (Dreamina web, CapCut Desktop, Jimeng, Doubao, ModelArk API): 30-second clips staged with explicit end states, integer-second timestamps, editing and extending finished footage, keyframes, storyboards, 3D clay-model references, and reference binding with roles and exclusions across up to 50 inputs. Use when someone wants a clip longer than 15s, more than 9 references, a prompt for Dreamina or CapCut, or wants to change something inside an already-generated shot. Triggers: 'seedance 2.5', 'thirty second clip', 'one-take', 'prompt for

AI, LLMs & Agents

Claude Recall

senapati484/claude-recall

Always-on Claude Code skill that bridges Claude with an Obsidian vault for persistent project memory. Automatically loads relevant context nodes from mindmap.json before every prompt (UserPromptSubmit hook) and saves structured session notes to the vault on exit (Stop hook). Zero manual invocation — hooks fire automatically. Context is AUTO-GENERATED using the claude CLI. Claude analyzes transcripts and project files to populate mindmap.json with stack, architecture decisions, gotchas, and current state. Users never need to manually edit. The /recall command lets users trigger on-demand contex

AI, LLMs & Agents

Polish Humanizer

waylooking-hub/polish-humanizer

Redaguje polskie teksty AI — sprawia, by brzmiały naturalnie i jak napisane przez człowieka, bez kalek angielskich, sztywnej składni i typowych „AI-tellsów" ChatGPT/Claude/Gemini. Używaj, gdy użytkownik mówi: „zhumanizuj", „brzmi jak chatgpt", „zrób naturalniej", „przepisz po ludzku", „brzmi sztucznie po polsku", „humanize / humanizer". NIE używaj do: obchodzenia detektorów AI; ukrywania autorstwa AI tam, gdzie disclosure jest wymagane (akademickie / dziennikarskie); tłumaczeń; pełnej stylistycznej humanizacji umów, regulaminów i dokumentów prawnych. Dla tekstów prawnych, urzędowych i technicz

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.