# Copilot Instructions for MCP Learning Project
<!-- Use this file to provide workspace-specific custom instructions to Copilot. For more details, visit https://code.visualstudio.com/docs/copilot/copilot-customization#_use-a-githubcopilotinstructionsmd-file -->
## Project Overview
This is an MCP (Model Context Protocol) server learning project built with TypeScript. The goal is to understand how to:
- Build MCP servers using the @modelcontextprotocol/sdk
- Create tools, resources, and prompts for AI assistants
- Integrate with Claude Desktop and other MCP clients
- Follow MCP best practices and patterns
## Development Guidelines
- Use TypeScript for all source code
- Follow the MCP SDK patterns and conventions
- Implement proper error handling and validation using Zod
- Use stdio transport for desktop integration
- Include comprehensive documentation and examples
## MCP Resources
You can find more info and examples at https://modelcontextprotocol.io/llms-full.txt
Additional reference: https://github.com/modelcontextprotocol/create-python-server
## Learning Focus Areas
1. **Tools**: Functions that can be called by LLMs (with user approval)
2. **Resources**: File-like data that can be read by clients
3. **Prompts**: Pre-written templates for specific tasks
4. **Transport**: Communication layer between server and client
5. **Integration**: Connecting with Claude Desktop and other clients
