gmail-mcp-server
Enables Claude to read and manage your Gmail account, including listing, searching, sending, drafting, replying to, archiving, labeling, and trashing emails via the Gmail API.
README
Gmail MCP Server
Local MCP server that lets Claude Desktop read and manage your Gmail (list, search, send, draft, reply, archive, labels, trash).
Claude asks → this server calls the Gmail API → actions run on your account.
Structure
gmail-mcp-server/
├── server.py # MCP tools + dispatch
├── gmail_auth.py # OAuth + token cache
├── gmail_tools.py # Gmail API wrappers
├── requirements.txt
├── credentials.json # from Google (do not commit)
└── token.json # created after first login (do not commit)
Setup
1. Install (Python 3.10+)
cd gmail-mcp-server
/opt/homebrew/bin/python3.12 -m venv .venv # or any python3.10+
source .venv/bin/activate
pip install -r requirements.txt
2. Google Cloud credentials
- Google Cloud Console → create/select a project.
- Enable Gmail API.
- OAuth consent screen → External → fill app info → add yourself as a Test user (keep status Testing).
- Credentials → Create → OAuth client ID → type Desktop app → download JSON.
- Save it as
credentials.jsonin this folder (must have an"installed"key, not"web").
3. Claude Desktop config
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"gmail": {
"command": "/Users/mdparwezansari/Desktop/gmail-mcp-server/.venv/bin/python",
"args": ["/Users/mdparwezansari/Desktop/gmail-mcp-server/server.py"]
}
}
}
Use absolute paths. Quit Claude fully (Cmd+Q) and reopen.
Claude starts the server automatically — you don’t need to run python server.py yourself.
4. First OAuth
In Claude, ask e.g. “List my recent emails.” Approve Google in the browser. That creates token.json for later runs.
Tools
| Tool | Description |
|---|---|
list_emails |
Recent mail (optional Gmail query) |
search_emails |
Search (from:, is:unread, etc.) |
get_email |
Full body by message ID |
send_email |
Send now |
create_draft |
Draft only |
reply_to_email |
Reply in-thread |
archive_email |
Remove from inbox |
mark_read / mark_unread |
Toggle unread |
delete_email |
Move to Trash |
list_labels / create_label |
Labels |
Scope: gmail.modify (change in gmail_auth.py if you want less access).
Optional env
GMAIL_CREDENTIALS_PATH— default./credentials.jsonGMAIL_TOKEN_PATH— default./token.json
Troubleshooting
| Problem | Fix |
|---|---|
mcp won’t install |
Need Python 3.10+ |
| Server not connected | Check paths; Cmd+Q Claude and reopen; see ~/Library/Logs/Claude/mcp-server-gmail.log |
| Access blocked | Add your Gmail as OAuth Test user |
| Re-login needed | Delete token.json and call a Gmail tool again |
Never commit credentials.json or token.json.
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.
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.
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.
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.