# Location AI Tools

**Location AI** tools from Mapbox bring geospatial capabilities to AI workflows — whether you're using an AI assistant to help you build applications, or building agentic applications that reason about location, routes, and places. These tools expose Mapbox maps, location data, and geospatial services through the [Model Context Protocol](https://modelcontextprotocol.io/introduction) (MCP), an open standard for connecting AI tools with external services.

## MCP Servers

Mapbox provides three MCP servers that give AI assistants direct access to Mapbox capabilities:

### [Mapbox MCP Server](https://docs.mapbox.com/ja/location-ai/mcp-servers/mcp-server/)

Exposes Mapbox [web services APIs](https://docs.mapbox.com/api/) in the Model Context Protocol. Use it to get directions, generate isochrones, geocode addresses, search for places, and render live interactive maps directly in your AI chat client.

### [Mapbox DevKit MCP Server](https://docs.mapbox.com/ja/location-ai/mcp-servers/devkit-mcp-server/)

Provides AI assistants with direct access to Mapbox developer APIs for style management, token creation, map feedback, and GeoJSON tooling. Ideal for developers building and iterating on Mapbox-powered applications with AI-assisted coding tools.

### [Mapbox Docs MCP Server](https://docs.mapbox.com/ja/location-ai/mcp-servers/docs-mcp-server/)

Gives AI assistants full-text search across all Mapbox documentation and the ability to retrieve complete documentation pages. No Mapbox access token required.

## Agent Skills

### [Mapbox Agent Skills](https://docs.mapbox.com/ja/location-ai/agent-skills/)

Pre-built knowledge modules that teach AI assistants best practices, integration patterns, and development workflows for building Mapbox applications across web, iOS, Android, and Flutter platforms. Agent Skills complement MCP servers by providing domain knowledge alongside actionable API capabilities.