Skip to main content
Tutorials

Snowplow CLI MCP tool setup

Data governance

  • Getting started with the Snowplow MCP server for tracking design

  • Installation and setup

  • Basic workflow

  • Next steps

Last updated on

Getting started with the Snowplow MCP server for tracking design

The Snowplow CLI MCP (Model Context Protocol) tool integrates Snowplow's data structure management capabilities directly into AI assistants like Claude. This enables natural language interaction for creating, validating, and managing your Snowplow tracking plans locally.

Important: The MCP tool creates and validates files on your local filesystem only. To sync changes to BDP Console, you'll use the regular CLI commands like snowplow-cli ds publish afterward.

What you'll learn​

  • How to set up the Snowplow CLI MCP tool with AI assistants like Claude Desktop, Cursor, or Copilot
  • Available MCP tools and their functions
  • Creating and validating data structures through conversation
  • AI-powered analysis for strategic tracking plan development

Demo: Using the Snowplow CLI MCP with Claude Desktop​

Prerequisites​

  • Snowplow CLI installed (installation guide)
  • Snowplow CLI configured with your BDP Console credentials (configuration guide)
  • Claude Desktop or another MCP-compatible client (Cursor or Copilot)
  • Filesystem access: If using Claude Desktop, you must run alongside an MCP filesystem server (e.g., @modelcontextprotocol/server-filesystem) to enable file operations. Other MCP clients (Cursor, Copilot, etc.) have filesystem access by default.

Available MCP tools​

The Snowplow CLI MCP server provides these tools:

Core tools​

  • get_context - Retrieves the built-in schema and rules that define how Snowplow data structures, data products, and source applications should be structured.
  • get_uuid - Generates valid v4 UUIDs required by many Snowplow components.

Validation tools​

  • validate_data_structures - Validates data structure files (events/entities). Must be called after creating or modifying any data structure.
  • validate_data_products - Validates data products and source applications. Must include both data product files AND their referenced source application files.
Last updated on