概览
What is MCP server for kintone?
MCP server for kintone is a Model Context Protocol server that connects AI tools such as Claude Desktop to the kintone business database platform. It allows AI assistants to explore and manipulate kintone data, enabling natural-language queries and updates without direct API calls. This server was community‑developed and is now superseded by an official Cybozu kintone MCP server.
How to use MCP server for kintone?
Download the latest executable from the release page and configure your MCP client (e.g., Claude Desktop) by adding a server entry with the required environment variables. Set KINTONE_BASE_URL (required) and either KINTONE_USERNAME/KINTONE_PASSWORD or KINTONE_API_TOKEN. Optionally restrict app access with KINTONE_ALLOW_APPS and KINTONE_DENY_APPS. Restart the client and start sending natural-language requests.
Key features of MCP server for kintone
- Connects AI tools to kintone data via MCP protocol
- Supports both password and API token authentication
- Allows read, query, and update of kintone records
- Optional allow/deny lists for app access control
- Standalone executable, no additional runtime required
Use cases of MCP server for kintone
- Query the latest status of a customer project in plain English
- Update a project’s progress percentage through conversation
- List all overdue tasks from a specified kintone app
- Ask AI to show records filtered by custom criteria
FAQ from MCP server for kintone
What authentication methods are supported?
You can authenticate using either a username/password pair or one or more API tokens. At least one method must be configured.
Can I restrict which kintone apps the server can access?
Yes. Use the KINTONE_ALLOW_APPS environment variable to list allowed app IDs, or use KINTONE_DENY_APPS to block specific apps. Deny takes priority over allow. By default all apps are allowed.
Is this the official kintone MCP server?
No. This is a community‑developed server. Cybozu has released an official kintone MCP server; this project will no longer receive new features.
What runtime or dependencies are required?
No additional runtime is needed. The server is distributed as a standalone executable for Windows, macOS, or Linux.
Where does the data live?
Data remains in your kintone environment. The server only acts as a bridge between the AI client and kintone’s APIs.
