MCP.so
登录

GRID MCP Server

@GRID-is

基本信息

分类

AI 与智能体

传输方式

stdio

作者

GRID-is

关于 GRID MCP Server

An MCP server for using GRID API directly from Claude for Desktop

GRID MCP Server 配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块,请前往代码仓库查看安装说明。

代码仓库

GRID MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is GRID MCP Server?

GRID MCP Server is a Model Context Protocol (MCP) server that enables you to query and interact with data in your GRID spreadsheets directly from Claude for Desktop using the GRID API.

How to use GRID MCP Server?

Clone the repository, install dependencies with npm install, then configure Claude for Desktop by adding the server to its claude_desktop_config.json with the correct absolute path to dist/index.js and your GRID API key in the GRID_API_TOKEN environment variable.

Key features of GRID MCP Server

  • Integrates GRID API into Claude for Desktop via MCP.
  • Allows natural‑language questions about spreadsheet data.
  • Requires Node.js, Git, and a GRID account with API key.
  • Supports local development with auto‑rebuild (npm run watch).
  • Debuggable using the MCP Inspector (npm run inspector).

Use cases of GRID MCP Server

  • Ask Claude questions about numbers and formulas in your uploaded spreadsheets.
  • Query workbook cells and ranges without leaving the Claude chat interface.
  • Automate spreadsheet analysis by combining Claude’s reasoning with GRID data.

常见问题

1

What prerequisites are needed to run GRID MCP Server?

You need Claude for Desktop, Node.js, Git, a GRID account with an uploaded spreadsheet, and a GRID API key from the GRID app settings.

2

How do I configure Claude for Desktop to use the server?

Edit Claude’s configuration file (`claude_desktop_config.json`) to add the server under `mcpServers` with the command `node`, the absolute path to `dist/index.js`, and your `GRID_API_TOKEN` set in the `env` object.

3

Where does my data live?

Your spreadsheet data remains in GRID; the server only sends queries via the GRID API. No data is stored locally by the MCP server.

4

How can I debug the server if something goes wrong?

Use the MCP Inspector by running `npm run inspector` in the project directory, which provides a URL with debugging tools for MCP servers over `stdio`.

评论

AI 与智能体 分类下的更多 MCP 服务器