Getting Started with Marco Polo
Set up your secure AI workspace in 5 minutes
Choose Your AI Client
Marco Polo works with any MCP-compatible AI client. Select the one you use:
Visit the home page for detailed installation instructions for your specific client.
Install the Marco Polo MCP Server
Add Marco Polo to your AI client using one of these methods:
🔗 Remote URL (Recommended)
https://mcp.marcopolo.dev
Copy this URL and add it as a remote MCP server in your client's settings.
📦 Claude Desktop Extension
Download the .mcpb file from the home page and double-click to install.
⚙️ Manual Configuration
Add this to your MCP configuration file:
{
"mcpServers": {
"marcopolo": {
"command": "npx",
"args": ["-y", "[email protected]", "https://mcp.marcopolo.dev"]
}
}
}
Authenticate Your Account
When you first connect to Marco Polo, you'll be prompted to authenticate:
Secure Authentication: Marco Polo uses OAuth 2.0. We never see or store your passwords. Your credentials are encrypted and stored securely.
Connect Your Data Sources
After authentication, configure your data connections:
🗄️ Databases
- PostgreSQL, MySQL, SQL Server
- MongoDB, Redis, Elasticsearch
- Oracle, DB2, Cassandra
☁️ Cloud Warehouses
- Snowflake, BigQuery
- Redshift, Databricks
- Azure Synapse, Athena
🔌 SaaS Applications
- Salesforce, HubSpot
- Stripe, Shopify
- Google Analytics, Mixpanel
📊 Data Platforms
- Looker, Tableau
- Mode, Metabase
- Amplitude, Segment
Connection Security: All credentials are encrypted at rest with AES-256. Connections use TLS 1.3 for data in transit. Your data never leaves your control.
Automatic Context Discovery: Once connected, Marco Polo scans your data sources to discover relationships, entities, and business concepts. The AI automatically receives structured context about your data—no manual documentation required.
Start Analyzing Data with AI
Your secure Kubernetes workspace is automatically provisioned, and Marco Polo has already scanned your data to understand its structure. Now you can start working with your data:
💬 Natural Language Queries
You ask:
"Show me revenue by product category this quarter"
The AI queries your database, processes results, and generates a report.
🔍 Cross-Database Analysis
You ask:
"Join customer data from PostgreSQL with orders from MongoDB"
The AI fetches data from both sources and uses DuckDB to join them.
📈 Data Visualization
You ask:
"Create a line chart showing monthly active users for the past year"
The AI queries your analytics platform and generates a visualization.
🤖 Complex Analysis
You ask:
"Calculate customer lifetime value and segment by cohort"
The AI writes and executes Python code to perform the analysis.