open-multi-agent/open-multi-agent is a TypeScript-native AI agent orchestration framework where a coordinator dynamically plans a task DAG at runtime and runs it on any LLM provider. Developers define high-level natural language goals instead of wiring nodes and edges up front.
Project overview
open-multi-agent/open-multi-agent is a TypeScript-native AI agent orchestration framework where a coordinator dynamically plans a task DAG at runtime and runs it on any LLM provider. Developers define high-level natural language goals instead of wiring nodes and edges up front.
Project type
AI Agent · Workflow
Use cases
Coding & Development · Automation
Deployment
Refer to project documentation
License
MIT
Best for
Developers and AI engineers who want a TypeScript-native library to dynamically plan a task DAG at runtime from high-level goals.
Key capabilities
A coordinator turns a goal into a task DAG at runtime, parallelizes independent tasks, and synthesizes the result for a deterministic scheduler.
Scaffolds production starters (PR Review Agent, Security Analysis Agent) or teaching DAG demos via 'npm create oma-app'.
Limitations and risks
Drops into any Node.js backend and is not a standalone app for end users.
Getting started
Install the core library via npm, import it into a Node.js backend, and define a natural language goal for the coordinator. TypeScript or Node.js knowledge is required.
Evidence and sources
GitHub project description: TypeScript AI agent orchestration framework with dynamic workflows. Describe the goal, not the graph: a coordinator plans the task DAG at runtime and runs it on any LLM (Claude, C…
README: `open-multi-agent` is an AI agent orchestration framework for TypeScript backends that drops into any Node.js app.
README: When creating a project, choose a **PR Review Agent**, **Security Analysis Agent**, or **multi-agent DAG starter demo**, then select a cloud/OpenAI-compatible provider or fully lo…
README: The short version: OMA is goal-driven, dynamic planning instead of rigid hand-wired graphs. Hand its Coordinator a goal and it builds the task DAG at runtime.
README: This is a monorepo. The published package is **`@open-multi-agent/core`**, and it lives in [`packages/core/`](packages/core/) — the source of truth for the library, its tests, exa…