# open-multi-agent-orchestration

A skill that teaches your agent to build AI agent teams on the open-multi-agent TypeScript framework: roles, models, task dependencies.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 710 installs · ★ 88
- Author: reason-machines
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/open-multi-agent-orchestration/SKILL.md
- Updated: 2026-10-09
- HTML: https://affpapa.org/en/nevibecode/skills/open-multi-agent-orchestration

## What it does

The skill hands your agent working knowledge of `open-multi-agent`, a TypeScript library for AI agent teams. Agents get separate roles, models and tool sets, while the framework resolves task order through a dependency graph, runs independent steps in parallel and keeps shared memory. Everything stays in one process, so there is no subprocess overhead.

It ships code examples: a single agent via `runAgent`, a team via `createTeam` and `runTeam`, and an explicit task pipeline via `runTasks` with the `dependsOn` field. Mixed teams are covered too, with one agent on Anthropic and another on OpenAI, plus custom tools through `defineTool()` and Zod schemas. Streaming output and progress events are in there as well.

You need a key before the first run: `ANTHROPIC_API_KEY` is required, `OPENAI_API_KEY` only if the team includes OpenAI models. The framework itself installs from npm as `@jackchen_me/open-multi-agent`. When a task fails, everything depending on it is skipped, so put checks on each graph node instead of only at the end.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill open-multi-agent-orchestration -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/open-multi-agent-orchestration/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/open-multi-agent-orchestration/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill open-multi-agent-orchestration -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/open-multi-agent-orchestration/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill open-multi-agent-orchestration -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/open-multi-agent-orchestration/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install reason-machines/trending-skills open-multi-agent-orchestration
```

2. **Manually: put the files in place**
- File: `.github/skills/open-multi-agent-orchestration/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill open-multi-agent-orchestration -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/reason-machines/trending-skills --path skills/open-multi-agent-orchestration
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/open-multi-agent-orchestration/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill open-multi-agent-orchestration -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/open-multi-agent-orchestration/SKILL.md

## When to use it
- build a team of agents with separate roles and models
- order tasks through a dependency graph
- run independent steps in parallel
- wire a custom tool with a Zod schema
- mix Anthropic and OpenAI models in one process

## Example prompts
- set up a team with an architect, a developer and a reviewer on open-multi-agent
- define a task pipeline with dependsOn: spec first, then implementation, then tests
- add a get_weather tool with defineTool and give it to the agent

## Similar
- [multi-agent-orchestration](https://affpapa.org/en/nevibecode/skills/multi-agent-orchestration.md) — skill · 2100 installs
- [skypilot-multi-cloud-orchestration](https://affpapa.org/en/nevibecode/skills/skypilot-multi-cloud-orchestration.md) — skill · 944 installs
- [orchestration](https://affpapa.org/en/nevibecode/skills/orchestration.md) — skill · 349800 installs
- [multi-agent-patterns](https://affpapa.org/en/nevibecode/skills/guanyang-multi-agent-patterns.md) — skill · 305 installs
- [Swarm Orchestration](https://affpapa.org/en/nevibecode/skills/swarm-orchestration.md) — skill · 2100 installs
- [agent-orchestration](https://affpapa.org/en/nevibecode/skills/agent-orchestration.md) — skill · 1400 installs

From the same set: [reason-machines/trending-skills](https://affpapa.org/en/nevibecode/by/reason-machines/trending-skills.md) — 49 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md