open-multi-agent
TypeScript multi-agent orchestration framework. Describe a goal, a coordinator decomposes it into a task DAG that runs on any LLM: Claude, ChatGPT, Gemini, DeepSeek, or local models.
Node.js backends that need a task graph built at runtime rather than hand-wired up front.
You would rather enumerate every node and edge yourself in a static graph.
About open-multi-agent
open-multi-agent is a multi-agent orchestration framework for TypeScript backends. Give it a goal; a coordinator agent decomposes it into a task DAG, parallelizes independents, and synthesizes the result. Drops into any Node.js backend.
Graph-first frameworks make you enumerate every node and edge up front. OMA runs a dynamic workflow: the task DAG is built at runtime, so it adapts to the goal instead of being hand-wired for one. The coordinator emits that plan as data for a deterministic scheduler to execute, so the plan is inspectable and replayable.
@open-multi-agent/core keeps a lightweight core. The orchestration engine plus the mainstream model providers (Anthropic, OpenAI, and any OpenAI-compatible endpoint) work out of the box; additional providers (Gemini, Bedrock), MCP, and the Vercel AI SDK bridge are opt-in peer dependencies you install only when you use them.
open-multi-agent is an open-source project written primarily in TypeScript, with 7.0k stars on GitHub. It was last updated in September 2026.
npm install @open-multi-agent/core
