open-mcp-client
by copilotkit
The open-mcp-client is a client application comprised of both a LangGraph agent and a CopilotKit frontend application designed to interact with MCP servers. It facilitates interaction with external tools through a LangGraph agent and provides a user interface for state synchronization.
Last updated: N/A
What is open-mcp-client?
The open-mcp-client is a client application that includes a LangGraph agent for interacting with MCP servers and a CopilotKit frontend for managing the user interface and state.
How to use open-mcp-client?
To use the open-mcp-client, first set up the required environment variables in both the root directory and the agent folder. Then, run either the frontend and agent separately using pnpm run dev-frontend and pnpm run dev-agent, respectively, or run both services together using pnpm run dev. Finally, open http://localhost:3000 in your browser. Ensure you have the correct API keys for Langsmith and OpenAI.
Key features of open-mcp-client
LangGraph agent for MCP server interaction
CopilotKit frontend for UI
State synchronization
Separate frontend and agent development
Environment variable configuration
Use cases of open-mcp-client
Interaction with MCP servers via agents
User interface for agent interaction
Agent state management
Connecting a user interface built in CopilotKit with agents running elsewhere
FAQ from open-mcp-client
What is LangGraph?
What is LangGraph?
LangGraph is a tool for building graph-based language models.
What is CopilotKit?
What is CopilotKit?
CopilotKit is a framework for building user interfaces that integrate with agents.
How do I set up the environment variables?
How do I set up the environment variables?
Create .env files in the root directory and the agent directory and add the required API keys.
How do I run the frontend and agent separately?
How do I run the frontend and agent separately?
Use the commands pnpm run dev-frontend and pnpm run dev-agent in separate terminals.
What should I do if I encounter errors?
What should I do if I encounter errors?
Check the logs for both the frontend and agent to identify the source of the error.