mcp-google-docs

# <img src="./assets/a1-logo.svg" alt="A1" width="40"> Google Docs MCP

**English** | [Русский](./README.ru.md)

[![npm](https://img.shields.io/npm/v/%40a1-x-tech%2Fmcp-google-docs)](https://www.npmjs.com/package/@a1-x-tech/mcp-google-docs)
[![Glama](https://glama.ai/mcp/servers/A1-x-Tech/mcp-google-docs/badges/score.svg)](https://glama.ai/mcp/servers/A1-x-Tech/mcp-google-docs)
[![CI](https://github.com/A1-x-Tech/mcp-google-docs/actions/workflows/ci.yml/badge.svg)](https://github.com/A1-x-Tech/mcp-google-docs/actions/workflows/ci.yml)
[![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](./LICENSE)

**A1 Google Docs MCP** lets an AI app read and edit Google Docs in plain language. Read a document as text or Markdown, change an exact passage, style headings, lists and tables, work through comment threads and export the result to PDF or DOCX.

It uses the Google Docs API with your Google account. It edits by exact index ranges rather than by guesswork, and makes the limits of the Docs API explicit instead of implying that every document task is possible.

- **27 tools.** Read a document as text, structure or Markdown, edit exact ranges, style characters and paragraphs, manage lists, tables, breaks, images and comment threads, and export to PDF, DOCX and more.
- **Connects from the conversation.** Say "connect Google Docs": the server walks you through the OAuth client, catches Google's redirect on `127.0.0.1` with PKCE and keeps the tokens itself — no config files, no restart.
- **Edits are surgical.** Changes address exact index ranges, and the server steers the assistant to re-read the document before every edit, because each change shifts the indexes after it.
- **Markdown both ways.** Create a document from Markdown or export to Markdown, PDF, DOCX and other formats; replacing a whole document with Markdown is a separate, explicitly destructive step.
- **No hidden Drive surface.** Export, Markdown conversion and comments use Drive endpoints internally, but the server exposes no general-purpose Drive tool.

Start with a read-only question:

> Read the launch plan document and summarize its unresolved comment threads.

[Connect the server](#quick-start) · [Explore use cases](#what-you-can-ask-it-to-do) · [Open technical documentation](#technical-documentation)

---

## See it work in a minute

> **You:** Show me the text and comments of the launch plan document.
>
> **Assistant:** Reads the document as compact text blocks and lists its comment threads. Nothing changes.
>
> **You:** Rewrite the “Timeline” paragraph to say the beta starts on March 3.
>
> **Assistant:** Shows the exact range it will replace and the proposed text, then asks for confirmation before editing.
>
> **You:** Confirm.
>
> **Assistant:** Replaces that one range. The rest of the document, its formatting and its comments stay as they were.

## Contents

- [Quick start](#quick-start)
- [What you can ask it to do](#what-you-can-ask-it-to-do)
- [How a document changes](#how-a-document-changes)
- [What can change](#what-can-change)
- [Getting access](#getting-access)
- [Configuration](#configuration)
- [Data, limits and background work](#data-limits-and-background-work)
- [Technical documentation](#technical-documentation)
- [Support](#support)

## Quick start

You need Node.js 20+ and a Google account. Credentials are not required at install time — the server connects from the conversation.

1. Add the server to your AI app.
2. Say "connect Google Docs": the assistant walks you through [creating the OAuth client and approving access](#getting-access) without editing config files.
3. Ask the read-only question above.

<details open>
<summary><strong>Codex</strong></summary>

<br>

**In the app:** open **Settings → MCP servers**, select **Add server**, choose **STDIO**, enter the command `npx -y @a1-x-tech/mcp-google-docs@latest` and environment variables `GOOGLE_DOCS_CLIENT_ID`, `GOOGLE_DOCS_CLIENT_SECRET`, `GOOGLE_DOCS_REFRESH_TOKEN`, then select **Save** and **Restart**.

**From the command line:**

```bash
codex mcp add google-docs \
  -- npx -y @a1-x-tech/mcp-google-docs@latest
```

```bash
codex mcp list
```

[Codex MCP documentation](https://learn.chatgpt.com/docs/extend/mcp?surface=cli)

</details>

<details>
<summary><strong>Claude Code</strong></summary>

<br>

```bash
claude mcp add \
  --transport stdio --scope user google-docs \
  -- npx -y @a1-x-tech/mcp-google-docs@latest
```

```bash
claude mcp list
```

[Claude Code MCP documentation](https://code.claude.com/docs/en/mcp)

</details>

<details>
<summary><strong>Claude Desktop</strong></summary>

<br>

The current official path is **Settings → Extensions**. For a custom desktop extension, open **Advanced settings → Extension Developer → Install Extension…**, select a `.mcpb` file and follow the prompts.

This repository currently publishes an npm stdio package and does not contain a `.mcpb` bundle. For Claude Desktop builds that still support local configuration, use the following JSON stdio configuration as a fallback:

```json
{
  "mcpServers": {
    "google-docs": {
      "command": "npx",
      "args": ["-y", "@a1-x-tech/mcp-google-docs@latest"]
    }
  }
}
```

In those builds, save it to `~/Library/Application Support/Claude/claude_desktop_config.json` on macOS or `%APPDATA%\Claude\claude_desktop_config.json` on Windows.

[Claude Desktop MCP documentation](https://support.claude.com/en/articles/10949351-getting-started-with-local-mcp-servers-on-claude-desktop)

</details>

<details>
<summary><strong>Cursor</strong></summary>

<br>

Add this to `~/.cursor/mcp.json` on macOS/Linux or `%USERPROFILE%\.cursor\mcp.json` on Windows:

```json
{
  "mcpServers": {
    "google-docs": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@a1-x-tech/mcp-google-docs@latest"]
    }
  }
}
```

[Cursor MCP documentation](https://cursor.com/docs/mcp)

</details>

<details>
<summary><strong>VS Code</strong></summary>

<br>

Run **MCP: Open User Configuration** and add:

```json
{
  "servers": {
    "google-docs": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@a1-x-tech/mcp-google-docs@latest"]
    }
  }
}
```

Check it with **MCP: List Servers**.

[VS Code MCP documentation](https://code.visualstudio.com/docs/agent-customization/mcp-servers)

</details>

## What you can ask it to do

### Read and export a document

- Read this document as text with headings and tables, and summarize it.
- Show the tab tree of the handbook document.
- Export the spec as Markdown; save the contract as a PDF file.

### Write and edit text

- Create a meeting-notes document from this Markdown.
- Insert a summary paragraph after the introduction.
- Replace every “Q3” with “Q4” across the document.
- Delete the outdated pricing section.

### Format and structure

- Turn these paragraphs into a numbered list; make this line a level-2 heading.
- Bold the key terms and link them to the glossary.
- Insert a 3×4 table for the roadmap and fill in the header row.
- Add a page break before the appendix; insert an image from a public URL.

### Work with comments

- List the open comment threads and summarize them.
- Reply to the comment about the deadline and mark it resolved.
- Add a comment quoting the sentence that needs legal review.

## How a document changes

1. `create_document` creates a **document** — empty, or converted from Markdown in one call.
2. Content is addressed by **indexes** — UTF-16 positions inside a tab’s body — and every insert or delete shifts all later indexes. The server steers the assistant to take fresh indexes from `read_document_text` before each edit and to edit from the end of the document backwards.
3. `import_markdown` replaces the **entire body**: comment anchors, positioned objects, headers/footers and extra tabs do not survive the conversion.
4. **Tabs** can be read and targeted, but the API cannot create, rename, delete or reorder them.
5. **Comments** live in Drive and are managed as threads. A new comment cannot be anchored to a text range — the anchor format is not public — so it attaches at document level, optionally quoting the text it refers to.

Exports are capped at 10 MB and do not include comments or suggestions. Inline images are fetched by Google from a public URL (PNG/JPEG/GIF, up to 50 MB and 25 megapixels); there is no channel for uploading image files.

## What can change

| Operation | What happens | Confirmation boundary |
|---|---|---|
| Read a document, its tabs or comments | Reads content and structure | No change |
| Export a document | Writes a local file when `output_path` is set; the document itself is untouched | Changes local files only |
| Create a document | Adds a new document | Changes Google Docs |
| Insert text, a table, a break or an image | Adds content | Changes a document |
| Style text or paragraphs, manage bullets | Overwrites formatting on a range | Changes a document |
| Replace or delete a range, find and replace | Removes existing content | Destructive |
| Replace the whole document with Markdown | Replaces the entire body | Destructive |
| Manage comments | Creates, replies, resolves or permanently deletes | Potentially destructive |
| Raw API request | Can call API methods without a dedicated tool | Potentially destructive |

The AI client controls confirmation prompts. The server marks reads, writes and destructive tools so the client can distinguish an inspection from a live change.

## Getting access

Google Docs requires OAuth 2.0; an API key is not enough. There are two ways in, and the first one needs no configuration files.

### Connect from the chat (recommended)

Say "connect Google Docs" and the assistant runs the flow with you:

1. `setup_instructions` prints the checklist: create or select a Google Cloud project, enable **Google Docs API**, configure the consent screen and create a **Desktop app** OAuth client.
2. Download that client's JSON ("Download JSON") and give the a