Server is distributed through PyPI package manager for installation
Provides a Python API for query execution and catalog services
Provides SQL query execution, table previewing, metadata catalog building, and dependency graph analysis for Snowflake data warehouses with built-in safety checks to prevent destructive operations
Igloo MCP - Snowflake MCP Server for Agentic Native Workflows
Igloo MCP is a standalone MCP server for Snowflake operations, designed for agentic native workflows with AI assistants. Built from the ground up with SnowCLI integration for maximum simplicity and performance.
✨ Features
🛡️ SQL Safety: Blocks destructive operations (DELETE, DROP, TRUNCATE) with safe alternatives
🧠 Intelligent Errors: Compact mode (default) saves 70% tokens; verbose mode for debugging
⏱️ Agent-Controlled Timeouts: Configure query timeouts per-request (1-3600s)
✅ MCP Protocol Compliant: Standard exception-based error handling
🚀 Zero Vendoring: Imports from upstream, stays in sync
📖 See Release Notes for details.
Available MCP Tools
Igloo MCP Tools
execute_query
- Execute SQL queries with safety checkspreview_table
- Preview table contentsbuild_catalog
- Build metadata catalogget_catalog_summary
- Get catalog overviewbuild_dependency_graph
- Build dependency graphtest_connection
- Test Snowflake connectionhealth_check
- Get system health status
See MCP Documentation for details.
Installation
For End Users (Recommended)
Install from PyPI for stable releases:
⚡ Quickstart
Success! 🎉 Your AI can now query Snowflake.
New to Snowflake? See Parameter Guide for help finding your account identifier and understanding which parameters are required.
Complete Setup Guide key-pair authentication
For production with key-pair authentication:
See Getting Started Guide for detailed setup instructions.
MCP Server (MCP-Only Interface)
Task | Command | Notes |
Start MCP server |
| For AI assistant integration |
Start with profile |
| Specify profile explicitly |
Configure |
| Interactive setup |
🐻❄️ MCP-Only Architecture Nanuk is MCP-only. All functionality is available through MCP tools.
Profile Selection Options:
Command flag:
igloo-mcp --profile PROFILE_NAME
(explicit)Environment variable:
export SNOWFLAKE_PROFILE=PROFILE_NAME
(session)Default profile: Set with
snow connection set-default PROFILE_NAME
(implicit)
Python API
Documentation
Examples
Query Execution via MCP
Data Lineage
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
Enables AI assistants to interact with Snowflake databases through SQL queries, table previews, and metadata operations. Features built-in safety checks that block destructive operations and intelligent error handling optimized for AI workflows.