mcpField
Q
Quip MCP Server

Quip MCP Server

Local
github.com/AvinashBoleProductivity6Updated 83w ago

About Quip MCP Server

Quip MCP Server connects Quip document operations to AI assistants through MCP. It supports reading and editing existing documents; document creation is currently unsupported.

Key features

  1. 01Read content by document ID
  2. 02Append content to a document
  3. 03Prepend content to a document
  4. 04Replace document content
  5. 05Document creation currently unsupported

Use cases

  1. 01View Quip documents from an AI assistant
  2. 02Append new content to an existing document
  3. 03Add content at the start of a document
  4. 04Update or replace document text

MCP support

The MCP server exposes Quip document tools for assistants, including reading, appending, prepending, and replacing content. Its document-creation tool is currently unsupported.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Quip MCP Server

A Model Context Protocol (MCP) server for Quip document operations that enables direct interaction with Quip documents from AI assistants like Claude.

Features

  • Read Documents: Fetch and display Quip document content by ID
  • Append Content: Add content to the end of existing documents
  • Prepend Content: Add content to the beginning of documents
  • Replace Content: Update document content
  • Create Documents: Intended support for creating new documents (currently redirects to web interface)

How It Works

This MCP server acts as a bridge between Claude and Quip documents. It works by:

  1. Receiving requests from Claude through the MCP protocol
  2. Executing a Python script (quip_edit_fixed.py) with the appropriate parameters
  3. Returning the results back to Claude

Prerequisites

  • Node.js v18 or higher
  • TypeScript
  • Python with quip library installed
  • A valid Quip access token

Installation

  1. Clone this repository:

    git clone https://github.com/AvinashBole/quip-mcp-server.git
    cd quip-mcp-server
    
  2. Install dependencies:

    npm install
    
  3. Build the project:

    npm run build
    
  4. Configure your MCP settings:

    {
      "mcpServers": {
        "quip": {
          "command": "node",
          "args": ["path/to/quip-server/build/index.js"],
          "env": {
            "QUIP_ACCESS_TOKEN": "your-quip-access-token",
            "QUIP_BASE_URL": "https://platform.quip.com"
          },
          "disabled": false,
          "autoApprove": []
        }
      }
    }
    

Usage

Once connected, the following MCP tools become available to Claude:

  • quip_read_document: Read a Quip document by its thread ID
  • quip_append_content: Append content to a document
  • quip_prepend_content: Add content to the beginning of a document
  • quip_replace_content: Replace document content
  • quip_create_document: Create a new document (currently unsupported)

Example usage in Claude:

<use_mcp_tool>
<server_name>quip</server_name>
<tool_name>quip_read_document</tool_name>
<arguments>
{
  "threadId": "YOUR_DOCUMENT_ID"
}
</arguments>
</use_mcp_tool>

Python Script Integration

The server expects a Python script called quip_edit_fixed.py in the path specified by the PYTHON_SCRIPT_PATH constant. This script should support the following operations:

  • read: Read document content
  • append: Add content to the end of a document
  • prepend: Add content to the beginning of a document
  • replace: Update document content

License

ISC License

Author

AvinashBole

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.