Sigma Computing MCP Server

Grant your AI agent (like Claude or Cursor) aggressive observational dominance over your Sigma Computing environment. The Sigma MCP equips your LLM to act as a fully autonomous data steward. Forget endlessly opening heavy BI platforms through browsers—now you can interrogate workbook metadata, map out Snowflake/BigQuery dependencies, and extract analytical taxonomies exclusively via natural conversational prompts interacting deeply with your dedicated API.

What you can do

• Massive Dashboard Espionage — Rip through your organizational analytics backbone via list_workbooks. Narrow down to specific layouts by drilling down structurally employing get_workbook_details and list_workbook_pages without leaving your console
• Lineage Cartography & Storage Maps — Trace the origin of datasets extracting organizational list_datasets and explicitly audit backend storage pipes mapping seamlessly back leveraging list_connections optimally
• Team Topology Surveillance — Interrogate user frameworks invoking list_organization_members cross-referential to rigid team structures invoking list_organization_teams instantly

How it works

1. Anchor this core interface directly within your native MCP agent framework
2. Safely entrench your SIGMA_CLIENT_ID and SIGMA_CLIENT_SECRET pairing inside the workspace to lock down security boundaries
3. Engage your agent pragmatically: "List all BI workbooks related to the North American region and see what underlying datasets they depend on!"