Mcp Server Kintone

What is Mcp Server Kintone about?

Mcp Server Kintone provides a Model Context Protocol (MCP) interface that lets AI assistants such as Claude Desktop query, update, and analyze data stored in kintone applications.

How to use Mcp Server Kintone?

Install – download the latest executable from the GitHub releases page and place it anywhere on your system.
Configure – edit the MCP client configuration (e.g., claude_desktop_config.json) and add a server entry that points to the executable. Set the required environment variables (KINTONE_BASE_URL, and either KINTONE_USERNAME/KINTONE_PASSWORD or KINTONE_API_TOKEN). Optional allow/deny lists can restrict access to specific app IDs.
Run – start the executable; the client will connect to the MCP server automatically.
Interact – ask your AI tool natural‑language questions or commands about kintone data.

Key features of Mcp Server Kintone

Direct MCP endpoint for kintone, compatible with any MCP‑aware AI tool.
Supports authentication via username/password or API tokens.
Fine‑grained access control using allow/deny app ID lists.
No additional runtime dependencies; a single binary is sufficient.
Works on Windows, macOS, and Linux.

Use cases of Mcp Server Kintone

Status queries – "What is the latest status of Customer A's project?"
Batch updates – "Update the progress of Project B to 50%."
Reporting – "Show me the projects that are behind schedule."
Workflow automation – integrate kintone data into broader AI‑driven processes.

FAQ from the Mcp Server Kintone

Q: Do I need both username/password and API token? A: No. Provide either a username/password pair or one or more API tokens.

Q: How do I restrict which apps the AI can access? A: Use KINTONE_ALLOW_APPS to list permitted app IDs, or KINTONE_DENY_APPS to explicitly block IDs (deny list takes precedence).

Q: What platforms are supported? A: The binary runs on Windows, macOS, and Linux. Place it anywhere and execute it directly.

Q: Do I need to restart my AI client after changing the config? A: Yes, restart the client (e.g., Claude Desktop) to pick up the new server settings.

Q: Can I run multiple MCP servers for different kintone domains? A: Yes. Create separate entries in the client config, each pointing to its own executable with distinct environment variables.

MCP server for kintone

Model Context Protocol (MCP) server for kintone. This server allows you to explore and manipulate kintone data using AI tools such as Claude Desktop!

Japanese version: 日本語の説明はREADME.ja.mdにあります。

Usage

1. Install

Download the latest release from the release page. You can place the executable file anywhere you like.

2. Configure MCP client like Claude Desktop

Configure your client to connect to the server.

For Claude Desktop, please edit file below:

MacOS/Linux: ~/Library/Application\ Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the following configuration to the mcpServers section:

{
  "mcpServers": {
    "kintone": {
      "command": "C:\\path\\to\\mcp-server-kintone.exe",
      "env": {
        "KINTONE_BASE_URL": "https://<domain>.cybozu.com",
        "KINTONE_USERNAME": "<your username>",
        "KINTONE_PASSWORD": "<your password>",
        "KINTONE_API_TOKEN": "<your api token>, <another api token>, ...",
        "KINTONE_ALLOW_APPS": "1, 2, 3, ...",
        "KINTONE_DENY_APPS": "4, 5, ..."
      }
    }
  }
}

Environment variables:

KINTONE_BASE_URL: (Required) The base URL of your kintone.
KINTONE_USERNAME: Your username for kintone.
KINTONE_PASSWORD: Your password for kintone.
KINTONE_API_TOKEN: Comma separated API token for kintone. You need to set either KINTONE_USERNAME and KINTONE_PASSWORD or KINTONE_API_TOKEN.
KINTONE_ALLOW_APPS: A comma-separated list of app IDs that you want to allow access. In default, all apps are allowed.
KINTONE_DENY_APPS: A comma-separated list of app IDs that you want to deny access. The deny has a higher priority than the allow.

You may need to restart Claude Desktop to apply the changes.

3. Start to use

Now you can interact with kintone using your own AI tools!

For example, you can say:

"What is the latest status of Customer A's project?"
"Update the progress of Project B to 50%."
"Show me the projects that are behind schedule."