MCP Config Validator

MCP is an open standard by Anthropic that defines how AI models connect to external tools, data sources, and services. An MCP server exposes capabilities (tools, resources, prompts) through a standardized JSON interface. Hosts like Claude Desktop, Cursor, and other AI tools act as MCP clients that connect to these servers.

MCP supports two transport types: 'stdio' (Standard I/O — the MCP server runs as a subprocess and communicates via stdin/stdout, most common for local tools) and 'sse' (Server-Sent Events — connects to a running HTTP server at a URL, used for remote/cloud MCP servers). The 'command' field is required for stdio; 'url' is required for sse.

Common reasons: (1) The 'command' or 'args' path is wrong — use absolute paths in production, (2) Missing required fields in the config, (3) JSON syntax errors, (4) The server process crashes on startup — check Claude Desktop logs at ~/Library/Logs/Claude/ (macOS). Use this validator to catch schema issues before troubleshooting runtime errors.

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json. Windows: %APPDATA%\Claude\claude_desktop_config.json. The top-level key is 'mcpServers' and each value is an MCP server config object. After editing, restart Claude Desktop to apply changes.

The 'env' object passes environment variables to the MCP server subprocess. Use it for API keys and secrets instead of hardcoding them in 'args'. For example: { "env": { "GITHUB_TOKEN": "ghp_xxxx" } }. Never commit configs with real tokens — use shell variable substitution or a secrets manager in CI/CD.