dbt-mcp and dbt-doctor
These are complementary tools: dbt-mcp provides the foundational protocol server for programmatic dbt interaction, while dbt-doctor adds AI-driven governance and quality checks as a specialized layer that can run alongside or through the base dbt-mcp interface.
About dbt-mcp
dbt-labs/dbt-mcp
A MCP (Model Context Protocol) server for interacting with dbt.
Exposes dbt project metadata and operations through 40+ tools across Discovery API, Semantic Layer, SQL execution, and dbt CLI capabilities—enabling AI agents to query lineage, model details, metrics, and trigger jobs. Connects to dbt Core, Fusion, and Platform environments, supporting both local manifest inspection and cloud-based operations with optional column-level lineage analysis via the Fusion engine.
About dbt-doctor
Astoriel/dbt-doctor
AI-driven quality & governance MCP Server for dbt projects. Audit coverage, profile data, detect schema drift, and auto-generate documentation — all through natural language with your AI assistant.
Implements the Model Context Protocol (MCP) to expose 12+ tools covering project auditing, data profiling via single-pass SQL queries, schema drift detection, and intelligent test suggestions based on column statistics. Operates as a read-only analysis layer with safe YAML writes using `ruamel.yaml` to preserve existing formatting and comments, integrating directly with Claude Desktop and Cursor via stdio transport and requiring only a compiled dbt manifest to function.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work