Marco Polo

Connect your data sources to AI with zero friction

Download and install the Marco Polo extension:

Download for Claude Desktop

Double-click the downloaded .mcpb file to install. Works on macOS, Windows, and Linux.

Alternative: Manual configuration

Or add this to your Claude Desktop configuration file:

{
  "mcpServers": {
    "marcopolo": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://prod01.prod.immersa.io/mcp"
      ]
    }
  }
}

Requires Node.js 18+ (including npx)

Config location:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Add Marco Polo to Claude Code via MCP:

claude mcp add --transport sse marcopolo https://prod01.prod.immersa.io/mcp

Requires Node.js 18+ and Claude Code CLI (npm install -g @anthropic-ai/claude-code)

One-click installation for VS Code:

Add to VS Code

This will open VS Code and prompt you to install the MCP server.

Add Marco Polo to Cursor's MCP configuration file:

{
  "mcpServers": {
    "marcopolo": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://prod01.prod.immersa.io/mcp"
      ]
    }
  }
}

After adding the server, restart Cursor to connect.

Requires Node.js 18+ (including npx)

Config file location:
macOS: ~/.cursor/mcp.json
Windows: %USERPROFILE%\.cursor\mcp.json
Linux: ~/.cursor/mcp.json

Add this to your Zed settings.json:

{
  "context_servers": {
    "marcopolo": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://prod01.prod.immersa.io/mcp"
      ]
    }
  }
}

Requires Node.js 18+ (including npx)

Access settings: Open Zed → Settings → Edit settings.json

With Developer Mode enabled, add Marco Polo as a remote MCP server in the Connectors tab:

https://prod01.prod.immersa.io/mcp

Enable Developer Mode in ChatGPT Settings:
Settings → Connectors → Advanced → Developer mode

Use this URL to connect to Marco Polo from any MCP-compatible client that supports streamable HTTP:

https://prod01.prod.immersa.io/mcp

Requires a client that supports remote MCP servers via streamable HTTP (SSE). Use with npx -y mcp-remote@latest or similar clients.

Connect to Any Data Source

50+ integrations out of the box. From cloud data warehouses to databases, analytics platforms to APIs.

Amazon Redshift Amazon Redshift
Snowflake Snowflake
Google BigQuery Google BigQuery
Databricks Databricks
PostgreSQL PostgreSQL
MySQL MySQL
Microsoft SQL Server Microsoft SQL Server
Oracle Oracle
MongoDB MongoDB
Elasticsearch Elasticsearch
ClickHouse ClickHouse
Presto Presto
Amazon Athena Amazon Athena
Google Analytics Google Analytics
Salesforce Salesforce
HubSpot HubSpot
Intercom Intercom
JIRA JIRA
Google Sheets Google Sheets
Prometheus Prometheus
InfluxDB InfluxDB
Cassandra Cassandra
DynamoDB DynamoDB
Azure Synapse Azure Synapse

Cloud Data Warehouses

Amazon Redshift Amazon Redshift
Snowflake Snowflake
Google BigQuery Google BigQuery
Databricks Databricks
Azure Synapse Azure Synapse
Azure SQL Database
Qubole
TreasureData

Databases

PostgreSQL PostgreSQL
MySQL MySQL
Microsoft SQL Server Microsoft SQL Server
Oracle Oracle
MongoDB MongoDB
SQLite
DB2
CockroachDB
MemSQL MemSQL
Vertica Vertica
Greenplum Greenplum

Big Data & Analytics

Amazon Athena Amazon Athena
Presto Presto
Hive Hive
Impala Impala
Apache Kylin
Druid Druid
ClickHouse ClickHouse
Elasticsearch Elasticsearch
Rockset Rockset

Time Series & NoSQL

InfluxDB InfluxDB
Prometheus Prometheus
Cassandra Cassandra
ScyllaDB ScyllaDB
Amazon DynamoDB Amazon DynamoDB
Amazon CloudWatch
CloudWatch Logs Insights
Graphite Graphite
Axibase Time Series

SaaS & APIs

Salesforce Salesforce
HubSpot HubSpot
Intercom Intercom
Google Analytics Google Analytics
Google Spreadsheets Google Spreadsheets
JIRA JIRA
Yandex Metrica Yandex Metrica
Yandex AppMetrica Yandex AppMetrica
JSON JSON
CSV
🔒

Enterprise-Grade Security

Secrets and OAuth tokens are stored by organization and encrypted using unique keys stored in KMS. Data and execution environments are airgapped for each user.

🎯

Smart Context Offloading

Stop wasting tokens on repetitive context. Marco Polo provides query examples and schema information exactly when needed, keeping your AI conversations focused and efficient.

Built-in Execution

Powered by DuckDB, Marco Polo provides a fast, embedded execution environment. Run queries, transform data, and get results without leaving your AI workflow.