manus-mcp
An MCP server that enables interaction with Google Drive through the Google Cloud API. It allows users to list, read, and upload files directly from MCP-compatible clients like manus.im or Cursor.
README
manus-mcp
MCP server on Fly.io that connects manus.im with Google Drive over Google Cloud.
- App URL: https://mcp-iwh3fa.fly.dev
- Repo: mitjaeichhorn/mcp
What it does
- OAuth: Users hit
/auth/google→ redirect to Google → callback at/auth/google/callback→ token stored on the server (in a Fly volume). - MCP tools: After one-time auth, Cursor/other MCP clients can call:
drive_list_files(folder_id, page_size)– list files in a Drive folderdrive_read_file(file_id)– download file content (text or base64)drive_upload_file(name, content, folder_id, mime_type)– upload a file
So manus.im (or any MCP client) can use this server as a bridge to Google Drive.
1. Google Cloud setup
- Go to Google Cloud Console and create or select a project.
- Enable the Drive API: APIs & Services → Library → search “Google Drive API” → Enable.
- OAuth consent screen: APIs & Services → OAuth consent screen → External (or Internal) → fill app name (e.g. “manus-mcp”), save.
- Credentials: APIs & Services → Credentials → Create credentials → OAuth client ID.
- Application type: Web application
- Name: e.g.
manus-mcp-fly - Authorized redirect URIs: add exactly:
https://mcp-iwh3fa.fly.dev/auth/google/callback
- Create → copy Client ID and Client secret.
2. Fly.io setup
-
Secrets (replace with your values):
fly secrets set GOOGLE_CLIENT_ID="your-client-id.apps.googleusercontent.com" fly secrets set GOOGLE_CLIENT_SECRET="your-client-secret"Optional (only if you use another URL):
fly secrets set BASE_URL="https://mcp-iwh3fa.fly.dev" -
Volume for token persistence (one-time, same region as app):
fly volumes create manus_mcp_data --region iad --size 1 -
Deploy:
fly deploy
3. First-time Google Drive connection
- Open: https://mcp-iwh3fa.fly.dev/auth/google
- Sign in with the Google account whose Drive you want to use.
- Approve the requested Drive access.
- You’ll be redirected back with “Connected. You can close this tab…”
- The server stores the refresh token in
/data(Fly volume). MCP tools can now call the Drive API.
4. Connecting manus.im (or any MCP client)
Point your MCP client at this server’s HTTP transport.
- URL:
https://mcp-iwh3fa.fly.dev - Transport: HTTP (FastMCP). If your client expects an MCP endpoint path, check its docs (e.g.
/mcpor root).
Example for a Cursor/IDE MCP config (format depends on your client):
{
"mcpServers": {
"manus-drive": {
"url": "https://mcp-iwh3fa.fly.dev"
}
}
}
After that, tools drive_list_files, drive_read_file, and drive_upload_file are available to the client (e.g. manus.im or Cursor).
Endpoints
| Path | Method | Description |
|---|---|---|
/ |
GET | 200 OK (for Fly checks) |
/health |
GET | JSON {"status":"healthy"} |
/auth/google |
GET | Redirect to Google OAuth |
/auth/google/callback |
GET | OAuth callback; saves token |
Local development
cd mcp
pip install -r requirements.txt
# No volume; token is stored in ./google_drive_token.json
export GOOGLE_CLIENT_ID="..."
export GOOGLE_CLIENT_SECRET="..."
export BASE_URL="http://localhost:8000" # or use ngrok and set BASE_URL to ngrok URL for callback
python server.py
Then open http://localhost:8000/auth/google (or your ngrok URL) to complete OAuth. Call the MCP tools via your client pointing at http://localhost:8000.
Summary
- manus.im ↔ Fly.io (mcp-iwh3fa.fly.dev) ↔ Google Drive via Google Cloud (OAuth + Drive API).
- One-time auth in the browser; tokens stored on Fly in a volume.
- MCP tools: list, read, upload files on the authenticated Drive.
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.
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.
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
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.
E2B
Using MCP to run code via e2b.