MCP.Market

Google Docs MCP server

by domdomegg·io.github.domdomegg/google-documents-mcp·v1.2.0·2 stars
No reviews yet

Allow AI systems to read and edit Google Docs.

B83/100grade B
Write a reviewInstall
What users say
No reviews yet
Be the first
Safety scan
B83/100

full report

Adoption
Growing

2 stars57 downloads/wk

Reviews

Write one

Nobody has reviewed Google Docs yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Google Docs tools (7, 3 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • document_append

    Append text to the end of a Google Doc. This is a convenience wrapper around batch_update that inserts text at the end of the document body.

  • document_batch_updatewrite action

    Apply one or more updates to a Google Doc. Supports inserting/deleting text, find/replace, inserting images and tables, managing bullets, and more. All requests are applied atomically - if any fails, none are applied.

  • document_createwrite action

    Create a new blank Google Doc with the specified title.

  • document_get_raw

    Get the full raw JSON structure of a Google Doc, including all tabs, formatting, headers, footers, and styles. Google Docs content is organized under tabs - body content is at tabs[].documentTab.body.content, not at the top level. Warning: responses can be very large. Consider using jq to extract specific fields, or use document_get_text for plain text content.

  • document_get_text

    Get the plain text content of a Google Doc without formatting, including all tabs. Google Docs content is organized under tabs (at tabs[].documentTab.body.content in the raw API), but this tool extracts and flattens the text for you. For very long documents, consider writing output to a file and reading specific sections.

  • document_insertwrite action

    Insert text at a specific location in a Google Doc. Use document_get to find the appropriate index. Index 1 is the beginning of the document body.

  • document_replace

    Find and replace all occurrences of text in a Google Doc. Returns the number of replacements made.

Public scan report

scanner v0.1.10 · 2026-09-28 · same rubric, same numbers if you re-run it

no findings
  • Code scan31 source files scanned; 31 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitystatic API keys via environment variables6/15
  • Maintenancelast push 19 days ago15/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 83/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Google Docs compares

Against every graded Productivity MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
83/100
higher than 60% of 1,301 Productivity servers
Tools
7
the typical Productivity server has 7
Tools that change things
43%
typical: 0%. Fewer write tools means less an agent can break
GitHub stars
2
more than 60% of 855 Productivity servers
Downloads a week
57
more than 42% of 433 Productivity servers with a package

What its tools let an agent do: 4 read · 3 change

Install Google Docs in Claude Code, Cursor or VS Code

Runs npx -y google-documents-mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add google-documents-mcp -- npx -y google-documents-mcp
Add to Cursor

What the publisher says

From the Google Docs repository's README, as published. We do not edit it. Read it on GitHub

google-documents-mcp

MCP server for Google Docs - read and edit documents.

Use Cases

Document Q&A: "What does the contract say about termination clauses?" → reads the doc and pulls out the relevant section.

Meeting-to-doc workflow: After a call, reads the Gemini transcript plus follow-up Slack discussion, then updates your team charter with the decisions made.

Release announcement: "Draft a blog post about the new feature based on the GitHub PRs and Slack discussions in #proj-awesome" → synthesizes technical changes into user-facing content.

Daily planning with persistent context: Your AI assistant reads a "planning log" doc at the start of each day, asks about yesterday's progress, helps you identify today's priority, then updates the doc with your plan. Context carries over across sessions - it remembers what you said you'd do and can follow up.

(These are just examples - any workflow that needs document reading or editing can use this. Use in combination with google-drive-mcp for finding files, deleting, comments, and sharing permissions.)

Setup

1. Create Google OAuth credentials

  1. Go to Google Cloud Console
  2. Create a new project (or use existing)
  3. Enable the Google Docs API
  4. Go to APIs & Services → OAuth consent screen, set up consent screen
  5. Go to APIs & Services → Credentials → Create Credentials → OAuth client ID
  6. Choose Web application
  7. Add http://localhost:3000/callback to Authorized redirect URIs
  8. Note your Client ID and Client Secret

2. Run the server

GOOGLE_CLIENT_ID='your-client-id' \
GOOGLE_CLIENT_SECRET='your-client-secret' \
MCP_TRANSPORT=http \
npm start

The server runs on http://localhost:3000 by default. Change with PORT=3001.

3. Add to your MCP client

With the server running, follow the instructions on install-mcp, which generates the right config for your MCP client (Claude Code, Claude Desktop, Cursor, Cline, VS Code, and more).

Architecture

This server acts as an OAuth proxy to Google:

graph LR
    A[MCP client] <--> B[google-documents-mcp] <--> C[Google OAuth/API]
  1. Server advertises itself as an OAuth authorization server via /.well-known/oauth-authorization-server
  2. /register returns the Google OAuth client credentials
  3. /authorize redirects to Google, encoding the client's callback URL in state
  4. /callback receives the code from Google and forwards to the client's callback
  5. /token proxies token requests to Google, injecting client credentials
  6. /mcp handles MCP requests, using the bearer token to call Google Docs API

The server holds no tokens or state - it just proxies OAuth to Google.

Tools

batch_update Request Types

The documentbatchupdate tool supports these operations:

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Google Docs: common questions

Is Google Docs MCP server safe?
Mostly: it is graded B (83/100). Read the Google Docs safety report
How do I install Google Docs?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Google Docs need an API key?
Yes. The registry entry asks for GOOGLE_ACCESS_TOKEN, GOOGLE_CLIENT_SECRET.
Is Google Docs maintained?
The last commit was 56 days ago (2026-09-09). The latest release is v1.2.0.
What can I use instead of Google Docs?
Servers from other publishers that do the same job: Agent-Native Content MCP server, Google Drive MCP server and Context7 MCP server.
How does Google Docs compare with other Productivity MCP servers?
safety score 83/100 (higher than 60% of 1,301 Productivity servers); tools 7 (the typical Productivity server has 7); tools that change things 43% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to Google Docs

Same job from other publishers: the closest match first, then the best rated.

  • Agent-Native Content
    Open-source Obsidian for MDX - edit local docs with agent assistance
    More from BuilderIO →
    A
  • Google Drive MCP
    Secure access to Google Drive, Docs, Sheets, Slides, and Calendar through MCP.
    B
  • Context7
    Up-to-date docs for AI agents
    A
  • Svelte MCP
    The official Svelte MCP server providing docs and autofixing tools for Svelte development
    A
  • Obsidian
    Read, write, search, and surgically edit Obsidian notes, tags, and frontmatter via MCP.
    A

More from domdomegg →