Support
A documentation site and component references are available, and the homepage offers demo booking and a Talk to an Engineer option.
Learn moreFrontend stack for connecting AI agents to users inside real apps

OpenBot by CopilotKit is a chat tool. CopilotKit, the team behind the OpenBot AI coworker event, is a self-hostable frontend stack that connects agents from many frameworks to users in web, mobile and chat apps, with generative UI, shared state and authentication.
OpenBot by CopilotKit comes from the team behind CopilotKit, a developer stack for placing AI agents inside real applications and the chat channels people already use. The OpenBot name refers to a live event on the homepage about building an open-source AI coworker, while the core product is the CopilotKit framework that such a coworker would be built on. The site positions it as an enterprise frontend stack that links any agent to any user, with the option to self-host. In practice, CopilotKit sits between an agent backend and the interface a person touches. It handles streaming and generative UI, user authentication, in-app actions with shared state, and what the site calls automatic learning from interactions. Developers wire an existing agent into a React, Vue, Angular, iOS or Android front end, or route the same agent into Slack, Teams and WhatsApp. The homepage lists compatibility with agent frameworks such as the OpenAI Agents SDK, Claude Agent SDK, LangChain, Google ADK, AWS Strands and Mastra, plus many more, and credits the AG-UI protocol as the layer connecting them. A refreshed Inspector is offered for examining agent behavior, and a copyable prompt lets a coding agent scaffold the setup inside a project folder. The listed use cases show the intended shape of the work: a company knowledge base that answers with citations and prepares next actions, an agent embedded in a SaaS product, bots in team messaging channels, and customer support with a clear human handoff. The audience is engineering teams building agentic features, not end users looking for a ready-made assistant. Among alternatives, it competes less with consumer chatbots and more with building a custom chat front end from scratch or adopting a single vendor's agent UI kit. Its pitch is framework neutrality and control over the interface. Pricing is not published on the homepage, and enterprise access runs through demo booking and conversations with an engineer.




A documentation site and component references are available, and the homepage offers demo booking and a Talk to an Engineer option.
Learn moreWorks with OpenAI Agents SDK, Claude Agent SDK, LangChain, Google ADK, AWS Strands, Mastra and many more agents. Front ends include React, Vue, Angular, iOS and Android; channels include Slack, Teams and WhatsApp.
Learn moreStreaming and generative UI, user authentication, in-app actions with shared state, automatic learning, a refreshed Inspector, a copyable agent setup prompt, and use case templates for knowledge bases, SaaS agents, messaging bots and support.
Learn more




