@cyanheads/who-gho-mcp-server
Query WHO Global Health Observatory data — 3,059 indicators across 194 member states with country, region, year, and sex filters via MCP.
README
<div align="center"> <h1>@cyanheads/who-gho-mcp-server</h1> <p><b>Query WHO Global Health Observatory data — 3,059 indicators across 194 member states with country, region, year, and sex filters via MCP. STDIO or Streamable HTTP.</b> <div>6 Tools • 2 Resources</div> </p> </div>
<div align="center">
</div>
<div align="center">
Public Hosted Server: https://who-gho.caseyjhand.com/mcp
</div>
Tools
6 tools for working with WHO Global Health Observatory data:
| Tool | Description |
|---|---|
who_search_indicators |
Search the GHO indicator catalog by keyword in indicator names |
who_list_indicators |
Browse the full indicator catalog with pagination |
who_get_indicator_metadata |
Fetch indicator names and supported filter dimensions for up to 10 codes |
who_list_dimensions |
List all dimension type codes available in the GHO API |
who_list_dimension_values |
List valid codes and labels for a dimension type (COUNTRY, REGION, SEX, etc.) |
who_query_indicator_data |
Query data rows for an indicator with spatial, temporal, and dimension filters |
who_search_indicators
Search the WHO GHO indicator catalog by keyword.
- Substring match on indicator names — try terms like
"life expectancy","immunization","mortality","diabetes", or"HIV" - Returns indicator codes and display names for use with
who_query_indicator_data - Reports total matches; suggests narrowing when the limit is reached
- Default limit 20, max 100
who_list_indicators
Browse the full indicator catalog with offset-based pagination.
- No keyword required — lists all 3,059+ indicators
- Pagination via
limit(default 50, max 500) andoffset - Returns
totalandhasMorefor iteration
who_get_indicator_metadata
Fetch metadata for one to ten indicator codes in a single call.
- Returns the full indicator name and the dimension types it supports (e.g.
COUNTRY,SEX,REGION,AGEGROUP) - Call before
who_query_indicator_datato confirm which filter dimensions are valid - Unresolved codes are reported in
notFoundrather than raising an error
who_list_dimensions
List all dimension type codes available in the GHO API.
- Returns every dimension type with its human-readable title
- Common types:
COUNTRY,REGION,SEX,WORLDBANKINCOMEGROUP,AGEGROUP - Use to discover codes before calling
who_list_dimension_values
who_list_dimension_values
List valid filter values for a single dimension type.
- Returns codes and labels for every value under the dimension (e.g. all 194 country ISO codes, all WHO region codes)
- Includes optional parent hierarchy fields (
parentCode,parentLabel,parentDimension) - Use to confirm exact codes before passing them to
who_query_indicator_data
who_query_indicator_data
Query data rows for a single WHO GHO indicator.
- Spatial filters (mutually exclusive):
country_codes(ISO 3166-1 alpha-3),region_codes(WHO regions), orincome_group_codes(World Bank groups) - Time range filter:
year_from/year_to - Sex filter:
SEX_BTSX(both),SEX_FMLE,SEX_MLE— only applies when the indicator uses SEX as its first cross-cutting dimension - Arbitrary
dim1_valuefor indicators using non-SEX cross-cutting dimensions - Optional uncertainty interval bounds (
low/high) viainclude_uncertainty(default true) - Default limit 200, max 1000; returns
totalRowsandtruncatedfor handling large result sets - Primary data tool in the find-then-query workflow
Resources
| Type | URI | Description |
|---|---|---|
| Resource | who://indicator/{indicatorCode}/metadata |
Indicator name and supported filter dimensions for a single code |
| Resource | who://dimension/{dimensionCode}/values |
All valid values for a dimension type |
Recommended workflow
who_search_indicators— find indicator codes by keywordwho_get_indicator_metadata— confirm which filter dimensions the indicator supportswho_query_indicator_data— fetch data with country/region/year/sex filters
To look up filter codes: who_list_dimensions → who_list_dimension_values.
Features
Built on @cyanheads/mcp-ts-core:
- Declarative tool definitions — single file per tool, framework handles registration and validation
- Unified error handling across all tools
- Pluggable auth (
none,jwt,oauth) - Swappable storage backends:
in-memory,filesystem,Supabase,Cloudflare KV/R2/D1 - Structured logging with optional OpenTelemetry tracing
- Runs locally (stdio/HTTP) or on Cloudflare Workers from the same codebase
WHO GHO-specific:
- Full coverage of the WHO GHO OData API v2 — indicators, dimensions, dimension values, and data queries
- Configurable base URL and request timeout for custom or mirrored deployments
- Parallel metadata fan-out for multi-code indicator lookups
Agent-friendly output:
- Tool descriptions encode the cross-tool workflow — agents discover the right call order from descriptions alone
- Structured truncation signaling (
truncated,truncatedNote,hasMore) so agents can decide whether to paginate - Discriminated error codes with
recoveryhints on every failure path
Getting started
Self-Hosted / Local
Add the following to your MCP client configuration file.
{
"mcpServers": {
"who-gho-mcp-server": {
"type": "stdio",
"command": "bunx",
"args": ["@cyanheads/who-gho-mcp-server@latest"],
"env": {
"MCP_TRANSPORT_TYPE": "stdio",
"MCP_LOG_LEVEL": "info"
}
}
}
}
Or with npx (no Bun required):
{
"mcpServers": {
"who-gho-mcp-server": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@cyanheads/who-gho-mcp-server@latest"],
"env": {
"MCP_TRANSPORT_TYPE": "stdio",
"MCP_LOG_LEVEL": "info"
}
}
}
}
Or with Docker:
{
"mcpServers": {
"who-gho-mcp-server": {
"type": "stdio",
"command": "docker",
"args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT_TYPE=stdio", "ghcr.io/cyanheads/who-gho-mcp-server:latest"]
}
}
}
For Streamable HTTP, set the transport and start the server:
MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# Server listens at http://localhost:3010/mcp
Prerequisites
- Bun v1.3.2 or higher (or Node.js ≥24).
- No API key required — the WHO GHO API is public.
Installation
- Clone the repository:
git clone https://github.com/cyanheads/who-gho-mcp-server.git
- Navigate into the directory:
cd who-gho-mcp-server
- Install dependencies:
bun install
Configuration
All configuration is validated at startup via Zod schemas in src/config/server-config.ts. Key environment variables:
| Variable | Description | Default |
|---|---|---|
MCP_TRANSPORT_TYPE |
Transport: stdio or http |
stdio |
MCP_HTTP_PORT |
HTTP server port | 3010 |
MCP_HTTP_ENDPOINT_PATH |
HTTP endpoint path where the MCP server is mounted | /mcp |
MCP_PUBLIC_URL |
Public origin override for TLS-terminating reverse-proxy deployments | none |
MCP_AUTH_MODE |
Authentication: none, jwt, or oauth |
none |
MCP_LOG_LEVEL |
Log level (debug, info, warning, error, etc.) |
info |
MCP_GC_PRESSURE_INTERVAL_MS |
Opt-in Bun-only forced-GC pressure loop (ms). Try 60000 if RSS grows under sustained HTTP load. |
0 (disabled) |
LOGS_DIR |
Directory for log files (Node.js only) | <project-root>/logs |
STORAGE_PROVIDER_TYPE |
Storage backend: in-memory, filesystem, supabase, cloudflare-kv/r2/d1 |
in-memory |
GHO_BASE_URL |
WHO GHO OData API base URL (override for custom/mirrored deployments) | https://ghoapi.azureedge.net/api/ |
GHO_REQUEST_TIMEOUT_MS |
HTTP request timeout in milliseconds | 30000 |
OTEL_ENABLED |
Enable OpenTelemetry | false |
Running the server
Local development
-
Build and run the production version:
# One-time build bun run rebuild # Run the built server bun run start:http # or bun run start:stdio -
Run checks and tests:
bun run devcheck # Lints, formats, type-checks, and more bun run test # Runs the test suite
Project structure
| Directory | Purpose |
|---|---|
src/mcp-server/tools |
Tool definitions (*.tool.ts). Six tools across indicator discovery, dimension lookup, and data queries. |
src/mcp-server/resources |
Resource definitions. Indicator metadata and dimension values resources. |
src/services/gho |
WHO GHO OData API service layer — HTTP client, query builder, types. |
src/config |
Server-specific environment variable parsing and validation with Zod. |
tests/ |
Unit and integration tests, mirroring the src/ structure. |
Development guide
See CLAUDE.md for development guidelines and architectural rules. The short version:
- Handlers throw, framework catches — no
try/catchin tool logic - Use
ctx.logfor logging,ctx.statefor storage - Register new tools and resources in the
createApp()arrays
Contributing
Issues and pull requests are welcome. Run checks and tests before submitting:
bun run devcheck
bun run test
License
This project is licensed under the Apache 2.0 License. See the LICENSE file for details.
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。