Sheet

Provides parsing, HTML conversion, and write-back for spreadsheet data via MCP clients.

python

3

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "yuqie6-mcp-sheet-parser-cot": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/MCP-Sheet-Parser-cot",
        "run",
        "main.py"
      ]
    }
  }
}

You can run the MCP Sheet Parser to let AI agents parse, convert, and update spreadsheet data. It communicates over standard input/output with MCP-compatible clients, enabling powerful tool-like interactions for XLSX, CSV, and other spreadsheet formats directly from your AI workflow.

How to use

You use the server by starting it as a local MCP service and connecting an MCP-compatible client such as Claude Desktop. The client sends requests to parse a sheet, convert sheets to HTML for visualization, or apply changes back to the original file. Perform actions by issuing natural language prompts to the AI agent, which then leverages the available tools to read, transform, and save spreadsheets.

How to install

Prerequisites: Python 3.8 or higher and a MCP-compatible runtime tool such as uvx.

Clone the project and install dependencies, then start the local MCP server.

# Step 1: Clone the project
git clone https://github.com/yuqie6/MCP-Sheet-Parser.git

# Step 2: Enter the project directory
cd MCP-Sheet-Parser

# Step 3: Install and sync dependencies using the recommended MCP runtime
uv sync

Additional setup and configuration

To connect the MCP server with an MCP client, add a local stdio server entry that points to the runtime and the main script. The following configuration uses uv to run the Python script from a local directory.

{
  "type": "stdio",
  "name": "sheet_parser",
  "command": "uv",
  "args": [
    "--directory",
    "/path/to/MCP-Sheet-Parser-cot",
    "run",
    "main.py"
  ]
}

Notes

The server exposes three core tools that cover the data processing workflow: parse_sheet, convert_to_html, and apply_changes. Use these tools to extract a structured view, visualize data as HTML, and persist edits back to the source file.

Security and best practices

Only expose locally to trusted clients. When handling sensitive spreadsheets, ensure proper access controls and consider backing up files before applying changes.

Available tools

parse_sheet

Parses a spreadsheet and returns a structured JSON representation. It can provide a quick overview with a limited preview by default, and return full data or styles if requested.

convert_to_html

Converts a spreadsheet to an HTML file while preserving much of the original styling so you can view the data in a browser.

apply_changes

Writes the modified data back to the original spreadsheet, creating a backup if requested, to persist AI-driven changes.