美國人口普查資料集搜尋 (ACS / decennial / PEP)

us_search_census_dataset

台灣專屬MIT

**關鍵字 / Keywords**: 美國人口普查 ACS 家戶調查人口統計十年普查 census demographic population household ACS decennial PEP geography FIPS Bureau vintage variable

Scope: US Census Bureau · api.census.gov dataset registry.

Search the 1,798-endpoint registry (ACS 1-year, ACS 5-year, decennial
PL/DHC/SF1, PEP annual, economic censuses, etc.) by keyword +
optional vintage + optional dataset-path filter.

Endpoint:
    https://api.census.gov/data.json (DCAT-shaped catalog)

Args:
    query: substring match on title + description + c_dataset path
           (case-insensitive). Empty string matches everything (with
           vintage / dataset filters still applied).
    vintage: exact vintage-year filter (e.g. 2022, "2020", 1999).
    dataset: substring match on the c_dataset path — common values:
             - "acs5" — ACS 5-year (block group down)
             - "acs1" — ACS 1-year (geos ≥65k population)
             - "dec"  — decennial (2020 / 2010 / 2000 / 1990)
             - "pep"  — Population Estimates Program (annual)
             - "ecn"  — Economic Census
             - "cbp"  — County Business Patterns
             - "nonemp" — Nonemployer Statistics
    limit: max rows returned (default 20, hard cap 200).

Returns dict with `results` — each row has `identifier`, `title`,
`description`, `vintage`, `dataset_path`, `base_url`,
`variables_url`, `geography_url`, `examples_url`. `base_url` feeds
`us_query_acs` / `us_query_decennial`. License: US Public Domain
(17 USC §105).

輸入 schema

{
  "properties": {
    "query": {
      "default": "",
      "title": "Query",
      "type": "string"
    },
    "vintage": {
      "anyOf": [
        {
          "type": "integer"
        },
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Vintage"
    },
    "dataset": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Dataset"
    },
    "limit": {
      "default": 20,
      "title": "Limit",
      "type": "integer"
    }
  },
  "title": "us_search_census_datasetArguments",
  "type": "object"
}

呼叫方式

配置好 Twinkle Hub 的 MCP client 後，agent 會看到 us_search_census_dataset。直接讓它呼叫即可，例如：

# Ask Claude / any MCP client:
請用 us_search_census_dataset 處理 "…"。

# It will call:
us_search_census_dataset(input="…")

尚未設定 Client？

Claude Desktop 約 3 分鐘即可完成設定 — 下載 .mcpb 檔案並雙擊即可，或參閱文件中的各 Client 安裝步驟。

查看使用者設定文件

輸入 schema

呼叫​方式

呼叫方式