CMS Data (data.cms.gov) MCP server for AI agents

Let Claude, ChatGPT or your own agent read CMS Data (data.cms.gov): 2 typed functions through one MCP server. No CMS Data (data.cms.gov) API key needed.

Functions
2
Sign-in
Not needed
Works with
MCP, HTTP, Node, Python

Connect

Just CMS Data (data.cms.gov)

An MCP server for CMS Data (data.cms.gov) alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.

https://api.bowmark.ai/mcp/site/data_cms
Every site

One MCP server for CMS Data (data.cms.gov) and every other site Bowmark covers. Your agent looks up what it needs and calls it.

https://api.bowmark.ai/mcp
Set it up in your client

Both need a Bowmark API key. The quickstart gets you one and a first result in a few minutes.

Functions

Open a function to see its arguments, what it returns and how to call it.

Read 2
  • getRowsRead rows from one data.cms.gov dataset (its latest version), optionally filtered by exact column values or a keyword, with the total number of…

    Arguments

    NameTypeRequired
    datasetIdstring
    SearchDatasets().datasets[].id, or a dataset's exact title.
    Yes
    limitnumber
    Default 25, max 500.
    No
    offsetnumberNo
    filtersRecord<string, string | number>
    Exact-match column filters using the dataset's own headers, e.g. { STATE: "TX", CITY: "AUSTIN" }.
    No
    keywordstring
    Free-text match across the row.
    No

    Returns data_cmsRowsResult

    interface data_cmsRowsResult {
      datasetId: string;
      /** rows as CMS publishes them; keys are the dataset's own column headers */
      rows: Record<string, unknown>[];
      /** rows matching the filters across the whole dataset */
      totalMatching: number | null;
      columns: string[];
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/data_cms/getRows

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.data_cms.getRows({
      datasetId: "f6f6505c-e8b0-4d57-b258-e2b94133aaf2",
      limit: 3,
      filters: { STATE: "TX" }
    });
  • searchDatasetsSearch data.cms.gov's dataset catalog by keyword; returns each dataset's id (for getRows), title, summary, latest version, update frequency and size.

    Arguments

    NameTypeRequired
    querystring
    Words to search the catalog for, e.g. "hospital enrollments".
    Yes
    limitnumber
    Default 10, max 50.
    No

    Returns data_cmsSearchResult

    interface data_cmsDataset {
      /** the uuid getRows takes; stable across versions */
      id: string;
      title: string;
      summary: string | null;
      version: string | null;
      lastUpdated: string | null;
      frequency: string | null;
      rowCount: number | null;
      columnCount: number | null;
      url: string | null;
      dataDictionaryUrl: string | null;
    }
    
    interface data_cmsSearchResult {
      datasets: data_cmsDataset[];
      totalMatches: number | null;
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/data_cms/searchDatasets

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.data_cms.searchDatasets({ query: "hospital enrollments", limit: 5 });

Use it from code