Google Docs MCP Proxy
Local stdio MCP server that handles Google OAuth2 locally and proxies authenticated requests to Google Docs and Google Drive REST APIs, enabling document and file operations from Cursor or other MCP clients.
README
Google Docs MCP Proxy
A local stdio MCP server that handles Google OAuth2 locally and proxies authenticated requests to the Google Docs and Google Drive REST APIs.
This solves the problem where Cursor cannot complete OAuth for Google's hosted Workspace MCP servers (Google does not support the cursor:// redirect protocol).
Architecture
Cursor Agent ──stdio JSON-RPC──► Local MCP Server ──Bearer──► docs.googleapis.com
│ └──► drive.googleapis.com
└──► OAuth2 Token Manager ──► Google OAuth
Prerequisites
-
A Google Cloud project with:
- Google Docs API enabled
- Google Drive API enabled
- OAuth consent screen configured with scopes:
https://www.googleapis.com/auth/documentshttps://www.googleapis.com/auth/drive
- OAuth 2.0 client of type Desktop app (not Web application)
— Desktop app clients allow
http://127.0.0.1redirects without pre-registration
-
Node.js 18+
Setup
1. Install dependencies
npm install
2. Configure environment
cp .env.example .env
Edit .env and fill in your OAuth client credentials from Google Cloud Console:
GOOGLE_CLIENT_ID=your-client-id.apps.googleusercontent.com
GOOGLE_CLIENT_SECRET=your-client-secret
3. Authorize (one-time)
npm run auth
This opens your browser to Google's consent screen, completes the OAuth2 flow, and saves tokens to token.json (gitignored). You only need to do this once — tokens refresh automatically.
Headless alternative: Set GOOGLE_REFRESH_TOKEN=<your-refresh-token> in .env to skip the interactive flow entirely.
4. Add to Cursor's mcp.json
Add the following entry to ~/.cursor/mcp.json (under mcpServers):
"google-docs": {
"command": "node",
"args": [
"D:\\Documents\\_Projects\\google-docs-mcp\\node_modules\\tsx\\dist\\cli.mjs",
"D:\\Documents\\_Projects\\google-docs-mcp\\src\\index.ts"
]
}
Restart Cursor after saving.
MCP Tools
google_docs_request
Make any request to the Google Docs REST API.
| Parameter | Type | Description |
|---|---|---|
path |
string | Path after /v1/, e.g. documents/{id}:batchUpdate |
method |
GET | POST | PATCH | PUT | DELETE | HTTP method |
body |
object (optional) | JSON request body |
query |
object (optional) | Query string params (string values) |
Examples:
Create a document:
{ "path": "documents", "method": "POST", "body": { "title": "My New Doc" } }
Read a document:
{ "path": "documents/DOC_ID", "method": "GET" }
Insert text via batchUpdate:
{
"path": "documents/DOC_ID:batchUpdate",
"method": "POST",
"body": {
"requests": [
{ "insertText": { "text": "Hello world\n", "location": { "index": 1 } } }
]
}
}
google_drive_request
Make any request to the Google Drive REST API v3.
Same parameters; base URL is https://www.googleapis.com/drive/v3/.
Examples:
List files:
{ "path": "files", "method": "GET", "query": { "pageSize": "10" } }
Search for docs:
{
"path": "files",
"method": "GET",
"query": { "q": "mimeType='application/vnd.google-apps.document'", "pageSize": "5" }
}
Development
npm run dev # tsx --watch (hot reload)
npm start # run once
npm run auth # re-authorize (if tokens are revoked)
Security notes
.envandtoken.jsonare gitignored — never commit them- OAuth credentials live in the project
.env, not in globalmcp.json - Token refresh happens automatically via native fetch; no manual intervention needed after initial auth
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.