Mappls Mcp

io.github.AmanMakesStuff由 GitHub 账号 AmanMakesStuff 发布v2.0.0更新于 Oct 24, 2026

MCP server for Mappls (MapMyIndia) — India-first maps API for AI agents

已验证STDIO仅桌面位置与生活网页搜索与抓取

概览

AI 生成的概览

一个 MCP 服务器，让助手获得面向印度的 Mappls（MapMyIndia）地理编码、路线规划、周边搜索和地图工具。

可以做什么
封装 Mappls（MapMyIndia）地图 API，提供以印度为先的位置数据。搜索与地理编码工具包括 geocode、reverse_geocode、autosuggest、text_search、nearby_search、place_details、address_analytics 和 validate_pincode。路线工具包括 get_directions、带实时路况的路线与距离矩阵、沿途兴趣点以及 snap_to_road。实用工具提供海拔、直线距离和静态地图图片链接。
适用场景
当助手需要准确的印度地址、邮编、地点搜索或印度境内的驾车与步行路线时适用。适合以印度为中心的位置问题，这类场景下通用全球地图服务器精度较低。
运行要求
通过 stdio 在本地运行，通常使用 npx mappls-mcp 或全局 npm 安装，因此需要 Node.js。需要在 MAPPLS_API_KEY 环境变量中提供 Mappls 静态 API 密钥，可从免费的 Mappls 开发者账号获取。需要访问 Mappls API 的网络连接。
安装前请注意
MAPPLS_API_KEY 密钥以明文形式写入客户端配置，请妥善保管配置文件。README 建议将 API 密钥的 IP 白名单字段留空以实现不受限访问，这会取消对密钥使用位置的限制。请求会把地址和坐标发送给第三方服务 Mappls，并受其免费额度限制。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Mappls Mcp，将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only，通过 STDIO。 STDIO 服务会启动本地进程，因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

mappls-mcp

An MCP (Model Context Protocol) server for Mappls — India's own maps platform (MapMyIndia). Gives AI agents accurate geocoding, routing, nearby search, and more for Indian addresses and locations.

Built because Google Maps MCP exists, but nothing existed for India-first location data.

Tools (18)

Search & Geocoding

ToolDescription
geocodeAddress or place name → lat/lng
reverse_geocodelat/lng → human-readable Indian address
autosuggestAutocomplete suggestions as user types
text_searchSearch by keyword or brand (e.g. "Starbucks Delhi")
nearby_searchFind hospitals, ATMs, petrol pumps, restaurants near a point
place_detailsFull details for a place using its eLoc code
address_analyticsStandardize a raw address into structured components
validate_pincodeValidate a 6-digit pincode, get district + state

Routing & Navigation

ToolDescription
get_directionsTurn-by-turn directions (driving/biking/walking)
get_directions_with_trafficTraffic-aware ETA using real-time conditions
distance_matrixDistances + durations between multiple points
distance_matrix_with_trafficTraffic-aware distance matrix
poi_along_routeFind fuel stations, restaurants, ATMs along a route
snap_to_roadSnap raw GPS coordinates to nearest road

Utilities

ToolDescription
elevationAltitude above sea level for any lat/lng point
aerial_distanceStraight-line (crow-fly) distance between two points
still_map_imageGenerate a static map image URL for any location

Getting Started

1. Get a Mappls API key (free)

  1. Sign up at about.mappls.com/api — free account, no credit card
  2. After login, click Create New Project → choose Cloud as the platform
  3. Open the project → go to the Credentials tab → copy the Static Key
  4. Go to the Whitelisting tab → leave the IP field blank for unrestricted access

The free tier includes generous API limits suitable for development and small projects.

2. Install

bash
npm install -g mappls-mcp

Or run directly without installing:

bash
npx mappls-mcp

Or clone and build locally:

bash
git clone https://github.com/AmanMakesStuff/mappls-mcpcd mappls-mcpnpm install && npm run build

3. Configure

Set your API key as an environment variable:

bash
export MAPPLS_API_KEY=your_static_key_here

Integration

Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

json
{  "mcpServers": {    "mappls": {      "command": "npx",      "args": ["mappls-mcp"],      "env": {        "MAPPLS_API_KEY": "your_static_key_here"      }    }  }}

Cursor / Windsurf

Add to your MCP settings:

json
{  "mappls": {    "command": "npx",    "args": ["mappls-mcp"],    "env": {      "MAPPLS_API_KEY": "your_static_key_here"    }  }}

Local build (alternative)

If you cloned the repo:

json
{  "mcpServers": {    "mappls": {      "command": "node",      "args": ["/absolute/path/to/mappls-mcp/build/index.js"],      "env": {        "MAPPLS_API_KEY": "your_static_key_here"      }    }  }}

Example Prompts

Once connected, ask your AI agent:

  • "Find the nearest hospital to Connaught Place, New Delhi"
  • "What's the address at 28.6139, 77.2090?"
  • "Get driving directions from Bandra to Andheri"
  • "Is pincode 400001 valid? What area is it?"
  • "How far is Delhi from Mumbai in a straight line?"
  • "What is the elevation of Shimla?"
  • "Find ATMs along the route from India Gate to Qutub Minar"
  • "Standardize this address: 237 Okhla Industrial Phase 3 ND 110020"
  • "Search for Domino's near Koramangala Bangalore"

Notes

  • All tools accept lat/lng in decimal degrees (e.g. 28.6139, 77.2090)
  • Routing tools use lat,lng format for input (the server handles coordinate conversion internally)
  • nearby_search uses plain keywords like "hospital", "atm", "restaurant" — Mappls maps these to category codes automatically
  • poi_along_route requires the encoded polyline from the geometry field of a get_directions response
  • still_map_image returns a URL — the image is a PNG map tile

License

MIT — built by AmanMakesStuff

来源：README.md，提交 37b4980

工具

0
SourceWeft 尚未在沙箱中启动过该软件包，因此其工具未知。

版本历史

1
  1. v2.0.0最新Sep 16, 2026

评分

已满足适用检查项的 40%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手；SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头，或保持端点能够响应探测，以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list，或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

    请在仓库或注册中心条目中声明许可证。