mcp-server

Open-Multi-Agent: AI Workflow Automation

Name: Open-Multi-Agent: AI Workflow Automation
Author: open-multi-agent

Get dynamic AI agent orchestration for your startup with open-multi-agent, used by founders across various niches.

6,841 stars2,431 forksTypeScriptGuide quality 8/10Updated 8/29/2026100% free · open source

What it does

Open-multi-agent is a TypeScript framework that allows dynamic AI agent orchestration with dynamic workflows, enabling the coordination of tasks across various large language models (LLMs) like Claude, ChatGPT, Gemini, DeepSeek, or local models

When to use it

•When you need to integrate multiple AI models into a single workflow
•When your startup requires dynamic task planning and execution using different LLMs
•When you want to run AI tasks on any available LLM, whether local or cloud-based

Ready-to-paste prompt

To run a task on the Claude LLM, use the following command: `npx ts-node runtime/index.ts --task my-task --model claude`

Heads up: Make sure to have the correct TypeScript version (14 or higher) installed, as open-multi-agent uses TypeScript features that are not compatible with lower versions

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

✅ Light setup

Installs with a command or two; your AI agent can do it for you.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/open-multi-agent && curl -fsSL https://workflowstacks.com/api/skills/open-multi-agent/claude-skill -o ~/.claude/skills/open-multi-agent/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How Open-Multi-Agent: AI Workflow Automation works

Codeflow

Free to inspect

Open-Multi-Agent: AI Workflow Automation is a very large TypeScript project (~70k lines across 278 code files, plus 182 test files). Setup is light: installs like a normal app. Reading the code is optional. Last commit this month, MIT license, has a test suite.

Size

Very large codebase

~70k lines · 278 code files · days to read — use, don't read

Setup

One-command install

Installs like a normal app. Reading the code is optional.

Runs on

Node.js

No API keys detected

TypeScript 99%JavaScript 1%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
package.json
Dependencies and the commands it exposes
4
packages/core/README.md
Inside packages/core/ — the main logic begins here

What's in each folder

packages/core/Core code — the actual logic458 files

docs/Documentation41 files

packages/create-oma-app/Package42 files

packages/release-bot/Package27 files

bench/Folder18 files

packages/otel/Package13 files

scripts/Helper scripts7 files

.github/CI / automation (GitHub Actions)28 files

READMEHas testsDocumentedCI checksMIT licenseUpdated this month

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

✅ Light setup

Installs with a command or two; your AI agent can do it for you.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/open-multi-agent && curl -fsSL https://workflowstacks.com/api/skills/open-multi-agent/claude-skill -o ~/.claude/skills/open-multi-agent/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

open-multi-agent

Language

TypeScript