Snowplow MCP

Snowplow's MCP server gives an assistant read and write access to the main areas of Snowplow Console, the control plane for a behavioral data pipeline. It covers data structures and schemas, letting an agent browse existing ones, create new ones, and search Iglu Central for reusable public schemas. Event specifications and tracking plans are exposed with their event volume metrics, alongside source applications and the entities attached to them. On the operations side, agents can inspect pipeline configuration, health metrics and collector settings, reach Snowplow Micro instances used for local testing, investigate failed events, manage data quality alerts, and view or update enrichment configurations. The data catalog is browsable and searchable, which matters in practice because it lets an agent check what is already being tracked before proposing a new schema and creating a duplicate. Signals workflows are covered as well, including attribute groups, services, interventions and publishing to the compute engine. A documentation tool serves Snowplow docs pages in the same session, so schema syntax and enrichment behavior can be checked without leaving the conversation. Built for analytics engineers and data teams whose real work is tracking plan design and failed event triage, and who would rather drive that from Claude Desktop, Claude Code, Cursor or Codex than from the console UI.