Baidu Maps Mcp

[中文](./README_zh.md)| English

[](https://github.com/baidu-maps/mcp)

[](LICENSE) [](https://pypi.org/project/mcp-server-baidu-maps/) [](https://www.npmjs.com/package/@baidumap/mcp-server-baidu-map)

🚀 Introduction

Baidu Map MCP Server is a fully MCP-compliant, open-source Location-Based Service (LBS) solution, providing a comprehensive suite of geospatial APIs and tools for developers and AI agents. As the first map service provider in China to support the Model Context Protocol (MCP), Baidu Map MCP Server bridges the gap between large language models (LLMs), AI agents, and real-world location data and services.

With Baidu Map MCP Server, you can easily empower your applications, LLMs, and agents with advanced mapping, geocoding, POI search, route planning, weather, traffic, and more — all via standardized, developer-friendly MCP interfaces.

Key Features:

• Full MCP Protocol Support: Seamless integration with any MCP-compliant agent, LLM, or platform.
• Rich LBS Capabilities: Geocoding, reverse geocoding, POI search, route planning (driving, walking, cycling, transit), weather, IP location, real-time traffic, and more.
• Cross-Platform SDKs: Official Python and TypeScript SDKs, easy CLI and cloud deployment.
• Enterprise-Grade Data: Powered by Baidu Maps' authoritative, up-to-date geospatial data.
• High Performance & Stability: Recommended SSE (Server-Sent Events) access for low latency and high reliability.
• Open Source & Extensible: MIT licensed, easy to customize and extend.

Whether you are building a travel assistant, logistics platform, smart city solution, or an LLM-powered agent, Baidu Map MCP Server provides the essential geospatial intelligence and tools you need.

The MCP Server architecture enables:

• Seamless AI Integration: Allows LLMs and agents to understand and process location data naturally
• Contextual Understanding: Provides rich geospatial context for more intelligent decision-making
• Standardized Interfaces: Consistent API design following MCP principles for easy integration
• Scalable Implementation: Suitable for projects of any size, from small applications to enterprise solutions

Whether you're building a navigation app, delivery service, smart city solution, or enhancing an AI agent with location awareness, Baidu Map MCP Server provides the tools and infrastructure you need to succeed.

🛠️ Supported Tools & APIs

Baidu Map MCP Server provides the following MCP-compliant APIs (tools):

| Tool Name | Description | |--------------------------|----------------------------------------------------------------------------------------------| | map_geocode | Convert address to geographic coordinates. | | map_reverse_geocode | Get address, region, and POI info from coordinates. | | map_search_places | Search for global POIs by keyword, type, region, or within a radius. | | map_place_details | Get detailed info for a POI by its unique ID. | | map_directions_matrix | Batch route planning for multiple origins/destinations (driving, walking, cycling). | | map_directions | Plan routes between two points (driving, walking, cycling, transit). | | map_weather | Query real-time and forecast weather by region or coordinates. | | map_ip_location | Locate city and coordinates by IP address. | | map_road_traffic | Query real-time traffic conditions for roads or regions. | | map_poi_extract* | Extract POI info from free text (requires advanced permission). |

> *Some advanced features require additional permissions. See [Authorization](#authorization) for details.

All APIs follow the MCP protocol and can be called from any MCP-compliant client, LLM, or agent platform.

⚡ Quick Start

1. Get Your API Key

Register and create a server-side API Key (AK) at Baidu Maps Open Platform. Be sure to enable “MCP (SSE)” service for best performance.

2. Python Integration

Install the SDK:

pip install mcp-server-baidu-maps

Run as a script:

python -m mcp_server_baidu_maps

Configure in your MCP client (e.g., Claude, Cursor):

{
  "mcpServers": {
    "baidu-maps": {
      "command": "python",
      "args": ["-m", "mcp_server_baidu_maps"],
      "env": {
        "BAIDU_MAPS_API_KEY": ""
      }
    }
  }
}

3. Node.js/TypeScript Integration

Install:

npm install @baidumap/mcp-server-baidu-map

Configure in your MCP client:

{
  "mcpServers": {
    "baidu-map": {
      "command": "npx",
      "args": [
        "-y",
        "@baidumap/mcp-server-baidu-map"
      ],
      "env": {
        "BAIDU_MAP_API_KEY": ""
      }
    }
  }
}

4. Recommended: Use SSE for low-latency, stable access

See SSE Quickstart.

5. More Platforms

• Claude/Agent/千帆AppBuilder: See [READMEzh.md](./READMEzh.md) for detailed integration guides and advanced configuration.

---

🚀 Advanced Use Cases

• Travel Planning Assistant:

Use map_search_places, map_directions, and map_weather to build an agent that plans optimal sightseeing routes, checks weather, and recommends POIs.

• Batch Route Matrix:

Use map_directions_matrix to calculate multiple routes and durations for logistics or delivery optimization.

• Text-to-POI Extraction:

Use map_poi_extract to extract POIs from user input or travel notes (requires advanced permission).

• Real-time Traffic & Weather-aware Navigation:

Combine map_road_traffic and map_weather for dynamic, context-aware travel suggestions.

• Integration with Claude, Qianfan, AppBuilder:

Seamlessly connect Baidu Map MCP Server to LLMs and agent frameworks for natural language geospatial reasoning.

See [READMEzh.md](./READMEzh.md) for more detailed Chinese documentation, configuration, and examples.

---

⛰️ Advanced Tutorials

• Geocoding API Guide
• POI Search API Guide
• Route Planning API Guide
• MCP Integration Guide
• SDK Development Guide

👩‍👩‍👧‍👦 Contributors

🌟 Star History

[](https://star-history.com/#baidu-maps/mcp&Date)

📄 License

[MIT](./LICENSE) © baidu-maps

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: baidu-maps
• Source: baidu-maps/mcp
• License: MIT
• Homepage: https://lbsyun.baidu.com/faq/api?title=mcpserver/base

Install and usage instructions live in the source repository linked above.