#265 · Primary category: AI Agents & Automation

open-multi-agent

agent-framework agent-orchestration agentic-ai ai-agents anthropic claude claude-code crewai-alternative deepseek dynamic-workflows gemini langgraph-alternative llm local-llm mcp multi-agent nodejs ollama openai typescript

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, ChatGPT, Gemini, DeepSeek, or local models).

Project last updated:08/29/26

GitHub Stars

6.8K

Forks

2.4K

Contributors

55

License

MIT

Why we included this project

Open Multi-Agent is a TypeScript orchestration framework for Node.js backends that want to go beyond single prompt-and-response calls. Instead of wiring a fixed chain of steps, a coordinator turns one written goal into a task graph at runtime, and a deterministic scheduler hands those tasks to a team of specialist agents and merges the results. That flexibility helps when the right order of operations only becomes clear once work has started. Every run stays inspectable: you can preview the plan before it dispatches, approve it, suspend it, replay it, resume from checkpoints after an interruption, and cap token or cost budgets. Because the same graph runs on Claude, OpenAI, Gemini, DeepSeek, or local models like Ollama, including CLI-based agents, you aren't locked to one provider when you orchestrate.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category