Frequently Asked Questions about dbt-docs-mcp Server

Q: What is the dbt-docs-mcp server? A: The dbt-docs-mcp (Model Context Protocol) server is a tool that provides a standardized way to interact with dbt project metadata, specifically dbt Docs artifacts (manifest.json, catalog.json). It allows querying node details, model/column lineage, and related metadata.

Q: What are the key functionalities of the dbt-docs-mcp server? A: Key functionalities include searching dbt nodes by name, column, or SQL code; inspecting node attributes; exploring lineage (upstream and downstream dependencies); and tracing column-level lineage (ancestors and descendants).

Q: What are the prerequisites for using the dbt-docs-mcp server? A: You need to have Python installed and uv for dependency management. You also need a dbt project with manifest.json and catalog.json files generated by dbt.

Q: How do I install and run the dbt-docs-mcp server? A: First, clone the repository. Then, optionally parse the dbt manifest for column-level lineage using the provided script. Finally, run the server using the mcp run command, providing the paths to your manifest and catalog files as environment variables.

Q: What is Model Context Protocol (MCP)? A: MCP is an open protocol that standardizes how applications provide context to LLMs. The dbt-docs-mcp server acts as a bridge, allowing AI models to access and interact with dbt project metadata.

Q: How does the dbt-docs-mcp server integrate with UBOS? A: The dbt-docs-mcp server can be integrated with UBOS to provide AI Agents with access to dbt project metadata. This enables agents to understand and interact with data pipelines, improving data quality, analysis, and governance.

Q: What are some potential extensions for the dbt-docs-mcp server? A: Suggested extensions include a tool for executing SQL queries, a tool for retrieving table/view/column metadata directly from the database, and a tool to search a knowledge base.

Q: What clients are compatible with the dbt-docs-mcp Server? A: Any MCP client that supports the mcp.json configuration format can connect to the dbt-docs-mcp Server. Examples include Claude desktop and Cursor.

Q: Why should I use dbt-docs-mcp on UBOS? A: UBOS provides a full-stack AI Agent development platform, allowing you to easily orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents with your LLM model, and create Multi-Agent Systems. The dbt-docs-mcp server on UBOS provides the critical data context layer for your AI Agents to understand and interact with your data pipelines.