cdc-health-mcp-server — Claude Code MCP server for CDC data
Lets your agent search the CDC open data catalog, read dataset schemas, run SoQL queries and pull mortality figures from CDC WONDER.
claude mcp add --transport http cdc-health-mcp-server https://cdc.caseyjhand.com/mcpWhat cdc-health-mcp-server does
The server sits between your agent and two CDC sources: the open data portal and CDC WONDER. Five tools come with it. One searches the dataset catalog, another returns a dataset's columns and row count, a third lists the catalog's categories and tags with entry counts, a fourth runs SoQL queries with filtering, grouping and full-text search, and the fifth queries WONDER for national mortality, population and death rates. Two resources mirror part of that: cdc://datasets for the most-viewed catalog entries and cdc://datasets/{datasetId} for one dataset's metadata. A prompt named analyze_health_trend walks through a health question step by step.
Transport is STDIO or Streamable HTTP. A public host at https://cdc.caseyjhand.com/mcp means you can skip running anything locally. The local route needs npx and a CDC app token passed as CDC_APP_TOKEN, alongside MCP_LOG_LEVEL, MCP_TRANSPORT_TYPE, MCP_HTTP_HOST and MCP_HTTP_PORT.
Watch the identifiers. cdc_get_dataset_schema expects a four-by-four datasetId such as bi63-dtpu, and charts or maps in the catalog are not tabular — querying one fails with not_queryable. Paging through rows without a unique tie-breaker in order gives unstable results, and the response tells you when that is the case.
What it helps with
- find a CDC dataset by keyword, category or tag
- inspect a dataset's columns and row count before querying
- pull mortality by year, sex and race from CDC WONDER
- filter and aggregate catalog data with SoQL
- list catalog categories and tags with entry counts
Example prompts for the agent
- Find CDC datasets about vaccination and show me their schemas
- Pull mortality by year and sex from the
underlying_2018_2024database - Count catalog entries per category in the CDC catalog
How to install cdc-health-mcp-server
1Platform CLI: remote server
claude mcp add --transport http cdc-health-mcp-server https://cdc.caseyjhand.com/mcp2Platform CLI
claude mcp add --env CDC_APP_TOKEN=<CDC_APP_TOKEN> --env MCP_LOG_LEVEL=<MCP_LOG_LEVEL> --env MCP_TRANSPORT_TYPE=<MCP_TRANSPORT_TYPE> --env MCP_HTTP_HOST=<MCP_HTTP_HOST> --env MCP_HTTP_PORT=<MCP_HTTP_PORT> --transport stdio cdc-health-mcp-server -- npx -y @cyanheads/cdc-health-mcp-serverReplace <…> with your own values. 2 more environment variables are in the server README
1Platform CLI: remote server
codex mcp add cdc-health-mcp-server --url https://cdc.caseyjhand.com/mcp2Platform CLI
codex mcp add cdc-health-mcp-server --env CDC_APP_TOKEN=<CDC_APP_TOKEN> --env MCP_LOG_LEVEL=<MCP_LOG_LEVEL> --env MCP_TRANSPORT_TYPE=<MCP_TRANSPORT_TYPE> --env MCP_HTTP_HOST=<MCP_HTTP_HOST> --env MCP_HTTP_PORT=<MCP_HTTP_PORT> -- npx -y @cyanheads/cdc-health-mcp-serverReplace <…> with your own values. 2 more environment variables are in the server README
Config file
{
"mcpServers": {
"cdc-health-mcp-server": {
"url": "https://cdc.caseyjhand.com/mcp"
}
}
}{
"mcpServers": {
"cdc-health-mcp-server": {
"command": "npx",
"args": [
"-y",
"@cyanheads/cdc-health-mcp-server"
],
"env": {
"CDC_APP_TOKEN": "<CDC_APP_TOKEN>",
"MCP_LOG_LEVEL": "<MCP_LOG_LEVEL>",
"MCP_TRANSPORT_TYPE": "<MCP_TRANSPORT_TYPE>",
"MCP_HTTP_HOST": "<MCP_HTTP_HOST>",
"MCP_HTTP_PORT": "<MCP_HTTP_PORT>"
}
}
}
}Replace <…> with your own values. 2 more environment variables are in the server README
Config file
{
"servers": {
"cdc-health-mcp-server": {
"type": "http",
"url": "https://cdc.caseyjhand.com/mcp"
}
}
}{
"servers": {
"cdc-health-mcp-server": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@cyanheads/cdc-health-mcp-server"
],
"env": {
"CDC_APP_TOKEN": "<CDC_APP_TOKEN>",
"MCP_LOG_LEVEL": "<MCP_LOG_LEVEL>",
"MCP_TRANSPORT_TYPE": "<MCP_TRANSPORT_TYPE>",
"MCP_HTTP_HOST": "<MCP_HTTP_HOST>",
"MCP_HTTP_PORT": "<MCP_HTTP_PORT>"
}
}
}
}Replace <…> with your own values. 2 more environment variables are in the server README
1Platform CLI: remote server
gemini mcp add -t http cdc-health-mcp-server https://cdc.caseyjhand.com/mcp2Platform CLI
gemini mcp add -e CDC_APP_TOKEN=<CDC_APP_TOKEN> -e MCP_LOG_LEVEL=<MCP_LOG_LEVEL> -e MCP_TRANSPORT_TYPE=<MCP_TRANSPORT_TYPE> -e MCP_HTTP_HOST=<MCP_HTTP_HOST> -e MCP_HTTP_PORT=<MCP_HTTP_PORT> cdc-health-mcp-server npx -y @cyanheads/cdc-health-mcp-serverReplace <…> with your own values. 2 more environment variables are in the server README
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about cdc-health-mcp-server
How do I install cdc-health-mcp-server in Claude Code?
For the hosted server: claude mcp add --transport http cdc-health-mcp-server https://cdc.caseyjhand.com/mcp. To run it locally: `claude mcp add --env CDC_APP_TOKEN=<CDC_APP_TOKEN> --env MCP_LOG_LEVEL=<MCP_LOG_LEVEL> --env MCP_TRANSPORT_TYPE=<MCP_TRANSPORT_TYPE> --env MCP_HTTP_HOST=<MCP_HTTP_HOST> --env MCP_HTTP_PORT=<MCP_HTTP_PORT> --transport stdio cdc-health-mcp-server -- npx -y @cyanheads/cdc-health-mcp-server`.
What does cdc-health-mcp-server do?
It searches the CDC open data catalog, returns dataset schemas and runs SoQL queries against them. A separate tool fetches mortality statistics from CDC WONDER across five databases.
Which CDC WONDER databases can I query?
Five of them: underlying_1999_2020 (the default), provisional, underlying_2018_2024, multiple_1999_2020 and multiple_2018_2024. Grouping by year, age group, sex or race works at the national level only.
