Mcp client example

What is Mcp client example?

Mcp client example is a sample CLI chat application that demonstrates an AI assistant using the Anthropic API with an MCP client and an MCP server. It is intended for developers learning to integrate MCP with Anthropic’s API.

How to use Mcp client example?

Set up by running pnpm install, then configure a .env file with ANTHROPIC_API_KEY and MCP_SERVER_SCRIPT_PATH. Build with pnpm build and run with pnpm start. The app connects to an MCP server via stdio and enters a chat loop until the user types ‘quit’.

Key features of Mcp client example

• CLI chat interface using Anthropic API.
• Connects to an MCP server over stdio transport.
• Automatically calls MCP tools when the AI requires them.
• Includes a sequence diagram for the full interaction flow.
• References a public time‑tools MCP server as an example.

Use cases of Mcp client example

• Learning how to build an MCP client with the Anthropic API.
• Prototyping a tool‑calling AI assistant.
• Testing custom MCP servers in a chat environment.
• Demonstrating the client‑server‑API flow for presentations or documentation.

FAQ from Mcp client example

What dependencies does Mcp client example require?

It requires Node.js, the Anthropic API key, and an MCP server script path (set in a .env file). The project uses pnpm for package management.

How does the client interact with the MCP server?

The client connects to the MCP server via stdio, retrieves tool definitions, and calls tools when the Anthropic API response indicates a tool should be used.

Where does the user’s API key and configuration live?

The API key and MCP server script path are stored in a local .env file. No data is sent elsewhere unless the MCP server handles it.