LangGraph is an open-source orchestration framework for building stateful AI agents, released under the MIT licence. Developers define agent workflows as graphs and gain durable execution (an agent resumes after a failure), memory within and across sessions, streaming and human-in-the-loop control, where people can inspect and modify an agent's state. LangGraph is a library (Python, plus a JavaScript version) that you install and run inside your own applications, and it can be used without LangChain. For debugging, evaluation and deployment, LangChain offers a separate platform called LangSmith.
The project is maintained by LangChain, Inc., which states that it is headquartered in the United States and is therefore subject to US jurisdiction. It is not a community-governed project. Because the code is open and runs locally, your organisation decides where agents run and where data is stored. If you choose LangChain's managed LangSmith services, that company's terms and jurisdiction apply. If your agent calls external model providers, their jurisdiction applies as well.