JP datasets search

jp_search_datasets

TW-specificMIT

Scope: Japan e-Gov catalog datasets.

Keyword search Japan e-Gov open-data catalog (18,150 datasets · FTS BM25).



Args:
    query: keyword(s) · 日本語 works best (e.g. "白書", "統計", "予算",
           "農林業センサス", "厚生労働省"). English tags also indexed.
    agency: filter by `organization_name` (e.g. "農林水産省")
    fmt: filter by resource format (e.g. "CSV", "XLSX", "PDF")
    limit: max results (default 20 · raise to 50+ when scanning for
           year variants or serial reports)

Returns dict with `results`, each row has: name, id, title, notes,
organization_name, organization_title, num_resources, tags, formats,
metadata_modified, score. `name` is what you pass to jp_list_resources.

Input schema

{
  "properties": {
    "query": {
      "title": "Query",
      "type": "string"
    },
    "agency": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Agency"
    },
    "fmt": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Fmt"
    },
    "limit": {
      "default": 20,
      "title": "Limit",
      "type": "integer"
    }
  },
  "required": [
    "query"
  ],
  "title": "jp_search_datasetsArguments",
  "type": "object"
}

How to call

Once your Twinkle Hub MCP client is configured, your agent will see jp_search_datasets. Just ask it to call — example:

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

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

Haven't set up a client yet?

Claude Desktop in 3 minutes — download the .mcpb and double-click, or see docs for other clients.

See user docs