MCP.so
ログイン

kintone MCP Server (Python3)

@r3-yamauchi

基本情報

カテゴリ

その他

トランスポート

stdio

作者

r3-yamauchi

kintone MCP Server について

概要はまだありません

kintone MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "kintone-mcp-server-python3": {
      "command": "uvx",
      "args": [
        "kintone-mcp-server-python3"
      ]
    }
  }
}

kintone MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is kintone MCP Server (Python3)?

kintone MCP Server (Python3) is a Model Context Protocol server that integrates with the kintone REST API. It allows AI assistants to fetch records from kintone apps, search for app information, and filter results using queries. The server supports both API token and password authentication, with automatic pagination for large datasets.

How to use kintone MCP Server (Python3)?

Install via uvx kintone-mcp-server-python3 (recommended) or pip install kintone-mcp-server-python3. Set environment variables in a .env file or in the MCP client configuration: KINTONE_SUBDOMAIN (required), KINTONE_AUTH_TYPE (api_token or password), and the corresponding token or credentials. Start the server with uvx kintone-mcp-server-python3 or python3 -m kintone_mcp_server_python3. The server exposes three tools: get_records, get_all_records, and get_apps, all invoked via JSON arguments.

Key features of kintone MCP Server (Python3)

  • Fetch records from kintone apps with pagination.
  • Search and retrieve kintone app information.
  • Supports API token and password authentication.
  • Automatic pagination handling for large record sets.
  • Query-based filtering of records.
  • All API requests use POST with X-HTTP-Method-Override.

Use cases of kintone MCP Server (Python3)

  • Retrieve specific records from a kintone app filtered by query.
  • Obtain all records from an app with automatic pagination.
  • Search for kintone apps by name, ID, code, or space ID.
  • Integrate kintone data into AI-driven workflows via MCP clients.

よくある質問

1

What authentication methods are supported?

The server supports two authentication types: API token authentication (`KINTONE_AUTH_TYPE=api_token`) and password authentication (`KINTONE_AUTH_TYPE=password`) with username and password.

2

How do I configure the server for an MCP client?

Set environment variables in the client configuration under `"env"`. The minimal required variable is `KINTONE_SUBDOMAIN`, plus either `KINTONE_API_TOKEN` or `KINTONE_USERNAME`/`KINTONE_PASSWORD`. Example configuration using `uvx` is provided in the README.

3

What tools does the server provide?

Three tools: `get_records` (paginated record fetch with limit/offset), `get_all_records` (fetch all records, auto‑paginated), and `get_apps` (search app info by name, IDs, codes, or space IDs).

4

Are there any limits on record fetching?

Yes. `get_records` has a default limit of 100 and a maximum of 500 records per call. `get_apps` has a default limit of 100 and a maximum of 100 apps. Both tools support offset for pagination.

5

What runtime and dependencies are required?

The server is a Python package installable via pip or uv. It requires Python 3 and the dependencies specified in the project (installed automatically). No specific Python version is mentioned beyond Python 3.

コメント

「その他」の他のコンテンツ