osm-tagging-schema

This MCP server provides a specialized bridge to the OpenStreetMap tagging knowledge base through the @openstreetmap/id-tagging-schema library. It exposes seven MCP tools across three categories: Tag Query, Preset Discovery, and Validation. You can query available tag values or search tags, discover and explore OSM presets with detailed configurations, and validate or suggest improvements for tags including checks for deprecated terms. The server is designed for integration with AI agents and LLM workflows, and is typically accessed via MCP clients such as Claude Code/Desktop or the MCP Inspector for testing. To use it, run the server through npx or via Docker, then configure your MCP client to connect to the osm-tagging-schema MCP server name and transport (stdio via CLI or HTTP/REST if supported by your client). For Claude Code/Desktop users, you can add the MCP server as an external tool with the following pattern, which mirrors the documented examples: npx -y @gander-tools/osm-tagging-schema-mcp. The MCP Inspector can be used to test all tools and inspect responses in an interactive UI, ensuring that each tool behaves as expected before integrating into your AI workflows.

Prerequisites:

• Node.js (recommended: latest LTS) and npm installed on your system
• Internet access to fetch the MCP package

Installation steps (preferred: using npx, no local installation required):

1. Ensure Node.js and npm are installed. Verify with: node -v npm -v

2. Run the MCP server directly without installing: npx -y @gander-tools/osm-tagging-schema-mcp

3. (Optional) Run via Docker for isolated environment: docker run -i ghcr.io/gander-tools/osm-tagging-schema-mcp:latest

4. (Optional) Verify accessibility with the MCP Inspector or your MCP client by pointing it to the osm-tagging-schema server configuration.

If you prefer to install locally for repeated use:

1. Install dependencies and build (if provided) in your project: npm install @gander-tools/osm-tagging-schema-mcp
2. Start the server from your project scripts, for example: node node_modules/@gander-tools/osm-tagging-schema-mcp/dist/index.js

Refer to the project docs for additional deployment options and troubleshooting tips.

Notes and tips:

• The server is production-ready and designed for integration with AI agents; it does not provide a user-facing UI by itself. Use the MCP Inspector for testing and debugging.
• The recommended quick-start method is via npx, which avoids local installation.
• Docker deployment is available and can run with stdio transport using docker run -i ... as shown in the README.
• For Claude Code/Desktop integration, configure the MCP server with the appropriate command and arguments (npx -y @gander-tools/osm-tagging-schema-mcp) and reference the documented tool names in conversations.
• You can customize the environment with standard environment variables if the MCP server exposes configuration knobs; there are no required environment variables listed in the README, but you may extend the env map as needed for your deployment environment.
• Use the Quick Start and Documentation links in the README to access tool references, configuration guides, usage examples, and deployment instructions.