OpenAgent is an open-source personal AI assistant that combines large language models, retrieval-augmented generation and agent loops in one self-hostable package. It can use the computer and the browser, run a coding agent, search a personal knowledge base and carry multi-step tasks through to completion. The project has gained more than 5,700 stars on GitHub and was actively developed in 2026, with a public demo available on its website.
OpenAgent follows the next-generation assistant pattern: a conversational interface backed by real tools. Behind the chat, agent loops plan and execute tasks, computer-use and browser-use modules operate software and websites, a coding agent handles programming work, and RAG connects the assistant to your own documents. Support for the Model Context Protocol lets it plug into external tools and services, and because it is self-hosted, your data and model choices stay under your control. The project positions itself as a general-purpose agent harness rather than a single-purpose chatbot: the same core can drive a research helper in the morning, automate browser chores at noon and work through a coding task in the afternoon. A live demo on the project website lets prospective users judge the experience before committing to a deployment.
Free. OpenAgent is open source under the Apache-2.0 license; you only pay for the models and infrastructure you choose to run it with.
Users and developers who want a capable, tool-using personal AI assistant that they host themselves instead of relying on a closed cloud service.