What problem does it solve? Connecting LLMs to external data sources and APIs requires correctly implementing MCP servers, configuring clients, and handling authentication, which is error-prone without a structured workflow. ## Core Features & Use Cases - MCP Server Implementation: Build custom MCP servers in TypeScript or Python with properly defined tool schemas and resource definitions. - Client Configuration: Configure MCP clients such as Claude Desktop and Cursor, including OAuth setup for services like Linear and GitHub. - Troubleshooting: Diagnose MCP connection, authentication, and token refresh issues with a structured testing workflow. - Use Case: You want Claude Desktop to create and list Linear issues. Use this Skill to build a Linear MCP server, configure OAuth, register it in the client config, and verify the tools respond correctly. ## Quick Start Create an MCP server that integrates the Linear API with tools for creating and listing issues, then configure it in Claude Desktop.