Project setup

Project Goal

• Building a CLI-based chatbot to understand how MCP clients and servers work together
• This is a learning project combining both client and server in one codebase (normally you'd build only one or the other)

Project Architecture

• Client: MCP client that connects to a custom MCP server
• Server: Custom MCP server with two tools implemented
• Tool to read document contents
• Tool to update document contents
• Data: Fake documents stored in memory (not persisted to disk)

Important Context

• In real projects, developers typically build either:
• An MCP server to distribute for others to use, OR
• An MCP client that connects to existing third-party MCP servers
• This project builds both to demonstrate how they interact

Setup Instructions

• Download the CLI_project. zip starter code
• Open project in code editor
• Follow setup directions in README. md file:
• Add API key to . env file
• Install dependencies (using UV or pip)
• Run the project:
• With UV: uv run main. py
• Without UV: python main. py
• A chat prompt will appear; test with a simple query like "what's one plus one"

Takeaways

• This is a learning project that combines both MCP client and server in one codebase to demonstrate how they work together
• The server implements two tools: reading document contents and updating document contents, with fake in-memory data
• In real projects, developers typically build either a server for others to use OR a client connecting to existing servers, but not both
• Setup requires adding an API key to . env, installing dependencies, and running with uv run main. py or python main. py
• Test the chatbot with simple queries after setup to verify the client-server interaction is working