`search_speeches`

Pack: kokkai-diet · Endpoint: https://gateway.pipeworx.io/kokkai-diet/mcp

Full-text search of Japanese National Diet (parliament) speeches back to 1947, via the National Diet Library’s kokkai.ndl.go.jp API. Requires “query” and/or “speaker”. Returns speaker, party (speaker_group), house, meeting, date, session, and a permalink for each hit — the actual speech text is a short preview here; fetch the full text with get_speech(speech_id). ${JP_FIRST_NOTE} PAGINATION: the API caps results at 100 per request (default 30) — check “next_start_record” in the response; if it is not null, more speeches exist beyond this page and were NOT returned. Never treat a page as the complete result set without checking it.

Parameters

Name	Type	Required	Description
`query`	string	no	Japanese keyword(s), space-separated for AND search, partial match. E.g. “半導体” (semiconductors), “少子化対策” (declining birthrate AND countermeasures). English keywords under-match — prefer Japanese.
`speaker`	string	no	Speaker name in Japanese, partial match, e.g. “岸田文雄”. OR-matched if it has multiple readings.
`speaker_group`	string	no	Political party / parliamentary group (会派) in Japanese, partial match, e.g. “立憲民主党”.
`house`	string	no	Chamber: “衆議院” (House of Representatives), “参議院” (House of Councillors), “両院” (joint), or “両院協議会” (joint committee).
`from`	string	no	Earliest speech date, inclusive, YYYY-MM-DD.
`until`	string	no	Latest speech date, inclusive, YYYY-MM-DD.
`start_record`	number	no	1-based offset for pagination (default 1). Use the previous response’s “next_start_record” to fetch the next page.
`max_records`	number	no	Results per request, 1-100 (default 30). The API rejects values outside this range.

Example call

Arguments

{
  "query": "半導体",
  "max_records": 5
}

curl

curl -X POST https://gateway.pipeworx.io/kokkai-diet/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_speeches","arguments":{"query":"半導体","max_records":5}}}'

TypeScript (@pipeworx/sdk)

import { Pipeworx } from '@pipeworx/sdk';
const pipeworx = new Pipeworx();

const result = await pipeworx.call('search_speeches', {
  "query": "半導体",
  "max_records": 5
});

More examples

{
  "speaker": "岸田文雄",
  "max_records": 5
}

Connect

Add this to your MCP client config, or use one-click install buttons:

{
  "mcpServers": {
    "kokkai-diet": {
      "url": "https://gateway.pipeworx.io/kokkai-diet/mcp"
    }
  }
}

See Getting Started for client-specific install steps.

search_speeches