ESCAP Data Explorer MCP Server
An MCP server that provides official statistical data for development indicators in Asia-Pacific. It wraps the ESCAP Data Explorer API, letting any MCP-compatible host query SDG and thematic data through natural language.
README
ESCAP Data Explorer MCP Server
An MCP (Model Context Protocol) server that provides official statistical data for development indicators in Asia-Pacific. It wraps the ESCAP Data Explorer API, letting any MCP-compatible host query SDG and thematic data through natural language.
Live endpoint:
https://dataexplorer.unescap.org/mcp/message
What is this?
The ESCAP Data Explorer is the official statistical platform of the United Nations Economic and Social Commission for Asia and the Pacific (ESCAP). It publishes hundreds of SDG and thematic indicators covering poverty, gender, environment, health, trade, and more — across 58 Asia-Pacific countries and territories.
This MCP server wraps the ESCAP Data Explorer's SDMX v2.1 REST API and exposes it as five structured tools that an AI agent can call in sequence to discover, validate, and retrieve statistical data — all through natural language.
Example use cases
- "What is the poverty headcount ratio in Thailand between 2010 and 2022?"
- "Show me SDG 5 gender equality indicators available for South Asia."
- "Compare CO₂ emissions per capita across ASEAN countries from 2000 to 2020."
- "Which Asia-Pacific countries have data on maternal mortality after 2015?"
- "Summarize progress on SDG 1 (No Poverty) for least developed countries."
Tools
The server exposes five tools that follow a guided workflow:
| Step | Tool | Description |
|---|---|---|
| 1 | get_indicators |
Browse the full SDG & Thematic indicator hierarchy. Returns indicator codes and descriptions. |
| 2 | get_indicator_metadata |
Fetch structural metadata (dimensions, attributes) for a chosen indicator in XML. |
| 3 | get_countries |
List all 58 reference areas with their codes and full names. |
| 4 | check_data_availability |
Confirm data exists for a given indicator, country, and year range before fetching. |
| 5 | get_data |
Retrieve observation data in SDMX format for a given indicator, country, and period. |
Quickstart — use the public endpoint
The server is already deployed and publicly accessible. No installation needed.
Claude Desktop
Install mcp-remote if you haven't already:
npm install -g mcp-remote
Add to your Claude Desktop config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"escap-data-explorer": {
"command": "mcp-remote",
"args": ["https://dataexplorer.unescap.org/mcp/message"]
}
}
}
Restart Claude Desktop. The five ESCAP tools will appear automatically.
Other MCP clients (native Streamable HTTP)
{
"mcpServers": {
"escap-data-explorer": {
"url": "https://dataexplorer.unescap.org/mcp/message"
}
}
}
Self-hosting
Requirements
- Node.js 18 or later
- npm 8 or later
- Docker + Docker Compose (for server deployment)
1. Clone the repository
git clone https://github.com/your-org/escap-mcp.git
cd escap-mcp
2. Install and build
npm install
npm run build
3. Run locally (stdio mode — for development)
node dist/index.js
Test with Claude Desktop by pointing to the local binary:
{
"mcpServers": {
"escap-data-explorer": {
"command": "node",
"args": ["/absolute/path/to/escap-mcp/dist/index.js"]
}
}
}
4. Deploy with Docker
Build the image:
docker build -t escap-mcp:latest .
docker compose up -d
Verify the server is running:
curl http://localhost:9000/health
# {"status":"ok","transport":"streamable-http","messagePath":"/message"}
Configuration
All configuration is via environment variables:
| Variable | Default | Description |
|---|---|---|
MCP_TRANSPORT |
stdio |
Transport mode: stdio for local dev, http for Docker/server |
PORT |
9000 |
HTTP port when MCP_TRANSPORT=http |
PATH_PREFIX |
"" |
Path prefix if your reverse proxy passes it through unchanged |
Reverse proxy (Apache)
If deploying behind Apache, replace your /mcp/ location block with the following. The key directive is flushpackets=on, which prevents Apache from buffering the streaming HTTP response:
<Location /mcp/>
ProxyPass http://localhost:9000/ flushpackets=on flushwait=1
ProxyPassReverse http://localhost:9000/
ProxyPreserveHost On
RequestHeader set X-Forwarded-Proto "https"
SetEnv no-gzip 1
SetEnv dont-vary 1
</Location>
Required modules: mod_proxy, mod_proxy_http, mod_headers.
sudo a2enmod proxy proxy_http headers
sudo systemctl reload httpd
API reference
This server connects to the ESCAP Data Explorer SDMX v2.1 REST API. The underlying endpoints used are:
| Purpose | Endpoint |
|---|---|
| Indicator hierarchy | https://dataexplorer.unescap.org/search/api/search?tenant=demo |
| Indicator metadata | https://api-dataexplorer.unescap.org/rest/v2/data/dataflow/ESCAP/{type}_Dataflow/2.10/ |
| Country codelist | https://api-dataexplorer.unescap.org/rest/codelist/escap/cl_ref_area |
| Data availability | https://api-dataexplorer.unescap.org/rest/availableconstraint/ESCAP,{type}_Dataflow,2.10/ |
| Data retrieval | https://api-dataexplorer.unescap.org/rest/data/ESCAP,{type}_Dataflow,2.10/ |
Dataset types are either SDG (Sustainable Development Goals) or Theme (thematic indicators).
Transport
The server implements the Streamable HTTP MCP transport (the current standard as of MCP SDK 1.x). Legacy SSE transport is not supported.
- Endpoint:
POST /message— all client-to-server communication - Session management: via
Mcp-Session-Idrequest/response header - Server push:
GET /messagewithMcp-Session-Idheader
Project structure
escap-mcp/
├── src/
│ └── index.ts # All server logic (single file)
├── dist/ # Compiled output (after npm run build)
├── Dockerfile # Two-stage Alpine build
├── docker-compose.yml # Production deployment
├── apache-mcp.conf # Apache reverse proxy config snippet
├── tsconfig.json
└── package.json
GitHub topic tags
Add these topics to your GitHub repository (Settings → Topics):
mcp mcp-server model-context-protocol asia-pacific statistics open-data sdmx unescap sustainable-development-goals sdg
Contributing
Contributions are welcome. Please open an issue first to discuss what you'd like to change.
License
MIT — see LICENSE for details.
About ESCAP
The United Nations Economic and Social Commission for Asia and the Pacific (ESCAP) is the regional development arm of the United Nations for the Asia-Pacific region. The ESCAP Data Explorer provides free, open access to official statistics from member states across economic, social, and environmental dimensions.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。