Cross-platform AI chat client that connects LLMs such as OpenAI, Claude, Ollama and DeepSeek to MCP servers.
CopilotKit/open-mcp-client
A TypeScript web app and LangGraph-based agent service for connecting MCP servers to an AI chat client, built on CopilotKit.
- Stars
- 1,651
- Forks
- 251
- Open issues
- 16
- Last commit
- 4 hours ago
What is it?
Open MCP Client is a starter-style web application for building an AI chat client that works with Model Context Protocol (MCP) tools. It is split into two parts: a frontend that provides the user interface and an agent service that handles the core logic.
The agent service is managed with Poetry and uses OpenAI and LangSmith API keys, which are configured through environment variables. Frontend and agent can be run in separate terminals for easier debugging, or together with a single pnpm command, and the app is served locally on port 3000.
The repository is a reference implementation rather than a packaged library, so it is most useful as a starting point for developers who want to see how an MCP-capable agent and chat UI fit together.
Key features
- 01Separate frontend and agent services for the UI and core logic
- 02Agent service managed with Poetry and configured via .env files
- 03Supports OpenAI and LangSmith API keys through environment variables
- 04Single pnpm command starts both services for development
Use cases
- Starting point for building a chat client that connects to MCP tools
- Learning how a frontend and a LangGraph-style agent service fit together
- Local experimentation with MCP-enabled AI agents during development
Quick start
- 1
Create the .env file at the project root and add your OpenAI and LangSmith API keys
touch .env - 2
Install Poetry to manage the agent service dependencies
pip install poetry - 3
Start the frontend and agent together
pnpm run dev - 4
Open http://localhost:3000 in your browser
Things to know
- Requires an OpenAI API key and a LangSmith API key to run.
- The repository metadata shows no license, although the README states MIT.
Topics
FAQ
- What is open-mcp-client?
- A TypeScript web app and LangGraph-based agent service for connecting MCP servers to an AI chat client, built on CopilotKit. Open MCP Client is a starter-style web application for building an AI chat client that works with Model Context Protocol (MCP) tools. It is split into two parts: a frontend that provides the user interface and an agent service that handles the core logic.
- Is open-mcp-client free to use? What license does it have?
- open-mcp-client has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
- Is open-mcp-client still maintained?
- Actively maintained. The last commit to open-mcp-client was on Oct 22, 2026.
- What can I use open-mcp-client for?
- Starting point for building a chat client that connects to MCP tools Learning how a frontend and a LangGraph-style agent service fit together Local experimentation with MCP-enabled AI agents during development
- How do I install open-mcp-client?
- Create the .env file at the project root and add your OpenAI and LangSmith API keys: touch .env Install Poetry to manage the agent service dependencies: pip install poetry Start the frontend and agent together: pnpm run dev Open http://localhost:3000 in your browser
- Where can I get open-mcp-client?
- The source code and installation instructions are on GitHub: https://github.com/CopilotKit/open-mcp-client
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A local-first AI chat client that connects to multiple LLM providers and adds productivity tools through the MCP protocol.
A locally deployed desktop AI chat client with MCP tool support, offering private conversation storage and connections to your own or third-party model APIs.
An AI agent that turns a URL into draft Twitter and LinkedIn posts, with human approval before publishing, built on LangGraph.
A VS Code extension that turns plain .chat.md files into editable AI chat sessions with MCP tool support across LLM providers.
Open-source long-horizon super agent harness that orchestrates sub-agents, memory, and sandboxes to research, code, and create across tasks lasting minutes to hours.