Google MCP Server
Enables appending text to Google Docs and creating Gmail drafts, with interactive terminal approval before each action.
README
Google MCP Server
A Python FastAPI server that exposes Google Docs and Gmail operations with interactive terminal approval before each action.
Features
- POST /append_to_doc — Append text to an existing Google Doc
- POST /create_email_draft — Create a Gmail draft email
- Interactive approval — Every action prints its name and payload to the terminal and waits for
ybefore proceeding - OAuth 2.0 — Authenticates once via browser; reuses
token.jsonon subsequent runs
Project Structure
google-mcp-server/
├── server.py # FastAPI app with tool endpoints
├── auth.py # Google OAuth authentication
├── docs_tool.py # Google Docs tool (append content)
├── gmail_tool.py # Gmail tool (create draft)
├── requirements.txt # Dependencies
├── README.md # This file
├── credentials.json # (NOT committed) OAuth client secrets from Google Cloud
└── token.json # (NOT committed) Auto-generated after first login
Prerequisites
- Python 3.10+
- A Google Cloud project with Google Docs API and Gmail API enabled
- OAuth 2.0 Desktop client credentials downloaded as
credentials.json
Google Cloud Setup
- Go to Google Cloud Console.
- Create or select a project.
- Enable these APIs:
- Configure the OAuth consent screen (External or Internal, add your email as a test user if External).
- Create credentials:
- APIs & Services → Credentials → Create Credentials → OAuth client ID
- Application type: Desktop app
- Download the JSON file and save it as
credentials.jsonin this directory.
Installation
cd google-mcp-server
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -r requirements.txt
Place your downloaded credentials.json in the google-mcp-server/ directory.
First-Time Authentication
On the first API call (or when token.json is missing/expired), a browser window opens for Google sign-in. After approval, credentials are saved to token.json and reused automatically.
To authenticate ahead of time:
python -c "from auth import get_credentials; get_credentials()"
Run the Server
python server.py
The server listens at http://127.0.0.1:8000. Keep the terminal visible — approval prompts appear there.
Interactive API docs: http://127.0.0.1:8000/docs
Usage Examples
Append to a Google Doc
Find the document ID in the URL: https://docs.google.com/document/d/<DOC_ID>/edit
curl -X POST http://127.0.0.1:8000/append_to_doc \
-H "Content-Type: application/json" \
-d '{"doc_id": "YOUR_DOC_ID", "content": "\nHello from the MCP server!\n"}'
The terminal shows:
============================================================
Action: append_to_doc
Payload: {'doc_id': '...', 'content': '...'}
============================================================
Approve? (y/n):
Type y to proceed, or anything else to reject (returns HTTP 403).
Create a Gmail Draft
curl -X POST http://127.0.0.1:8000/create_email_draft \
-H "Content-Type: application/json" \
-d '{"to": "recipient@example.com", "subject": "Hello", "body": "Draft created via API."}'
Approve in the terminal when prompted. The draft appears in Gmail under Drafts.
OAuth Scopes
| Scope | Purpose |
|---|---|
https://www.googleapis.com/auth/documents |
Read and modify Google Docs |
https://www.googleapis.com/auth/gmail.compose |
Create and manage Gmail drafts |
Security Notes
- Never commit
credentials.jsonortoken.json(both are in.gitignore). - The approval gate is a local safety check — run the server only on trusted machines.
- Revoke access anytime at Google Account Permissions.
Troubleshooting
| Issue | Fix |
|---|---|
FileNotFoundError: Missing credentials.json |
Download OAuth desktop credentials from Google Cloud Console |
403 access_denied during OAuth |
Add your Google account as a test user on the OAuth consent screen |
403 Action rejected by user |
You declined the prompt; retry and type y |
| Token expired | Delete token.json and re-authenticate, or let the server refresh automatically |
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。