Gmail MCP Server
Enables AI agents to search, read, send, and organize Gmail emails via MCP protocol.
README
<p align="center"> <h1 align="center">📧 Gmail MCP Server</h1> <p align="center"> A powerful Gmail MCP server for AI agents — search, read, send, and organize emails via Claude Desktop, Claude Code, Cursor, and more. </p> </p>
<p align="center"> <a href="https://www.npmjs.com/package/gmail-mcp-server"><img src="https://img.shields.io/npm/v/gmail-mcp-server?color=%2300d4aa&style=flat-square" alt="npm version" /></a> <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue?style=flat-square" alt="MIT License" /></a> <a href="https://nodejs.org"><img src="https://img.shields.io/badge/node-%3E%3D18-brightgreen?style=flat-square" alt="Node.js >= 18" /></a> <a href="https://modelcontextprotocol.io"><img src="https://img.shields.io/badge/MCP-compatible-purple?style=flat-square" alt="MCP Compatible" /></a> </p>
<p align="center"> <strong>⭐ If you find this project useful, please consider giving it a star on GitHub! It helps others discover it. ⭐</strong> </p>
⚡ Quick Start
# One-command setup — auto-configures your AI client
npx gmail-mcp-server setup
That's it. The setup wizard will:
- Check your Google OAuth credentials
- Let you choose your AI client (Claude Desktop, Claude Code, Cursor, Windsurf)
- Auto-write the config file
- Run the OAuth authentication flow
✨ Why choose this MCP?
While Google provides workspace integrations and generic email readers exist, this MCP is explicitly built for AI Agents acting autonomously:
- Zero-Config Setup:
npx gmail-mcp-server setupdetects your OS and AI client, automatically editing your local configuration JSON. No manual path mapping required. - Built for AI Agents: Includes tools like
advanced_filter(search + bulk actions with LLM-friendly dry-runs),batch_modify_emails(modify 1000 emails at once), andget_frequent_contacts(turn an inbox into a CRM context). - 100% Local & Private: Bring-your-own-credentials. Tokens remain strictly on your machine. No proxy servers, no telemetry, and no third-party email routing.
- Modular & Extensible: Clean, deeply typed domain-driven structure (e.g.,
src/tools/compose.ts,src/tools/search.ts) making it extremely easy for developers to fork and extend.
🔧 Tools (20 tools)
| Category | Tool | Description |
|---|---|---|
| Search | search_emails |
Search with Gmail query syntax (from:, is:unread, subject:, etc.) |
| Read | get_email |
Full email content — body, headers, attachments |
get_thread |
All messages in a thread | |
| Compose | send_email |
Send a new email |
reply_email |
Reply (or reply-all) keeping the thread | |
forward_email |
Forward with optional note | |
create_draft |
Save a draft (optionally in a thread) | |
| Organize | modify_email |
Add/remove labels, mark read/unread |
batch_modify_emails |
Bulk modify up to 1000 emails | |
trash_email |
Move to trash | |
untrash_email |
Restore from trash | |
batch_trash_emails |
Bulk trash | |
mark_spam |
Mark as spam | |
advanced_filter |
Search + bulk action (archive, label, etc.) with dry-run | |
| Labels | list_labels |
List all labels with IDs |
create_label |
Create new label (supports nesting) | |
delete_label |
Delete user-created label | |
| Extract | extract_addresses |
Extract unique addresses from search results |
format_emails |
Filter, sort, reformat email data | |
| Profile | get_profile |
Your email address, message/thread counts |
get_frequent_contacts |
Most-contacted addresses from sent mail | |
| Attachments | get_attachment |
Download attachment content |
🚀 Step-by-Step Setup Guide
Follow these steps to get your Gmail MCP server running in less than 5 minutes.
1. Create Google Cloud Project & Enable API
- Open the Google Cloud Console.
- Create a New Project: Click the project dropdown (top left) → "New Project" → Name it
Gmail MCP→ "Create". - Enable Gmail API: Search for "Gmail API" in the search bar → Click it → Click Enable.
2. Configure OAuth Consent Screen
- Go to APIs & Services → OAuth consent screen.
- Select User Type: "External" (or "Internal" if you have a Google Workspace) → Click Create.
- App Information: Fill in "App name" (e.g.,
Gmail MCP) and "User support email". - Developer Contact: Fill in your email address.
- Click Save and Continue until you reach the Summary.
- Add Test Users: Go back to the "OAuth consent screen" tab → Under "Test users" → Click + ADD USERS → Enter your Gmail address → Click Save.
3. Generate Credentials
- Go to APIs & Services → Credentials.
- Click + CREATE CREDENTIALS → Select OAuth client ID.
- Application type: Select Desktop app.
- Name: Give it a name (e.g.,
Gmail MCP Client). - Click Create.
- Download JSON: In the "OAuth 2.0 Client IDs" list, click the download icon (↓) for your new client.
- Rename & Move: Save this file as
credentials.jsonin your project folder (or keep it ready for the setup wizard).
4. Run the Setup Wizard
Open your terminal and run:
npx gmail-mcp-server setup
The wizard will guide you through:
- Finding your
credentials.json. - Selecting your AI client (Claude, Cursor, etc.).
- Authorizing the app via your browser.
🖥️ Client Configuration
Claude Desktop
npx gmail-mcp-server setup # Select "Claude Desktop"
Or manually add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"gmail": {
"command": "npx",
"args": ["-y", "gmail-mcp-server"]
}
}
}
Claude Code
npx gmail-mcp-server setup # Select "Claude Code"
Or add to ~/.claude/mcp.json:
{
"mcpServers": {
"gmail": {
"command": "npx",
"args": ["-y", "gmail-mcp-server"]
}
}
}
Cursor
npx gmail-mcp-server setup # Select "Cursor"
Or add to .cursor/mcp.json:
{
"mcpServers": {
"gmail": {
"command": "npx",
"args": ["-y", "gmail-mcp-server"]
}
}
}
Windsurf
npx gmail-mcp-server setup # Select "Windsurf"
🛠️ CLI Commands
npx gmail-mcp-server setup # Interactive setup wizard
npx gmail-mcp-server auth # Run OAuth flow
npx gmail-mcp-server serve # Start MCP server (default)
npx gmail-mcp-server doctor # Diagnose issues
npx gmail-mcp-server --help # Show help
npx gmail-mcp-server --version # Show version
🩺 Troubleshooting
If you run into issues, try the following:
- Run the Doctor:
npx gmail-mcp-server doctorwill diagnose common problems with your credentials, tokens, and system. - Missing
credentials.json: Ensure you've downloaded the file from Google Cloud Console and it's named exactlycredentials.jsonin your current directory. - Invalid Grant / Expired Token: If the server can't connect, try running
npx gmail-mcp-server authto reset your OAuth tokens. - Port 3000 Busy: The auth flow uses port 3000. If it's busy, the flow might fail. Ensure no other apps are using port 3000 during setup.
💬 Usage Examples
Once configured, ask your AI agent:
- "Search my emails for unread messages from GitHub"
- "Get the full content of email ID 18abc123"
- "Show me the whole thread for this conversation"
- "Reply to that email saying thanks"
- "Forward the invoice to accounting@company.com"
- "Create a draft reply for that message"
- "Mark all newsletters from last month as read"
- "Extract all email addresses from my last 50 emails"
- "Who do I email most frequently?"
- "Download the PDF attachment from that email"
🔒 Security
credentials.json— never committed (in.gitignore)- Tokens stored in
~/.gmail-mcp/token.json— outside the project, auto-refreshed - OAuth scope:
gmail.modify(read + modify, no permanent delete or admin) - All data stays local — no external servers, no telemetry
🌍 Environment Variables
| Variable | Default | Description |
|---|---|---|
GMAIL_CREDENTIALS_PATH |
./credentials.json |
Path to your OAuth credentials file |
GMAIL_MCP_DIR |
process.cwd() |
Base directory for credentials lookup |
🤝 Contributing
We welcome contributions! See CONTRIBUTING.md for guidelines.
git clone https://github.com/neutral-Stage/gmail-mcp-server.git
cd gmail-mcp-server
npm install
npm run build
📄 License
MIT — use it however you want.
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.