mattijsdp/dbt-docs-mcp

Overview

An MCP server that enables interaction with dbt Docs through the Model Context Protocol. It allows MCP clients to query and explore dbt documentation programmatically.

Best for

Best for
Developers using dbt who want to query their documentation through MCP-compatible assistants or tools

Use cases

• Querying dbt model metadata and lineage from documentation
• Integrating dbt Docs into AI-assisted development workflows
• Automating documentation lookups via MCP-compatible tools

How to use

Tools exposed

• uv
• git
• python

Tested with

Claude Desktop, Cursor

Example client config

{\n    "mcpServers": {\n        "DBT Docs MCP": {\n        "command": "uv",\n        "args": [\n            "run",\n            "--with",\n            "networkx,mcp[cli],rapidfuzz,dbt-core,python-decouple,sqlglot,tqdm",\n            "mcp",\n            "run",\n            "/Users/mattijs/repos/dbt-docs-mcp/src/mcp_server.py"\n        ],\n        "env": {\n            "MANIFEST_PATH": "/Users/mattijs/repos/dbt-docs-mcp/inputs/manifest.json",\n            "SCHEMA_MAPPING_PATH": "/Users/mattijs/repos/dbt-docs-mcp/outputs/schema_mapping.json",\n            "MANIFEST_CL_PATH": "/Users/mattijs/repos/dbt-docs-mcp/outputs/manifest_column_lineage.json"\n        }\n        }\n    }\n}

Notes

An MCP server that enables interaction with dbt Docs through the Model Context Protocol. It allows MCP clients to query and explore dbt documentation programmatically.

23 stars on GitHub. Last updated 2025-08-19. Licensed MIT.

Use cases

• Querying dbt model metadata and lineage from documentation
• Integrating dbt Docs into AI-assisted development workflows
• Automating documentation lookups via MCP-compatible tools

Pros

• Leverages the standard MCP protocol for broad compatibility
• Lightweight Python implementation easy to extend
• Directly connects to existing dbt Docs output

Cons

• Small community and limited adoption (23 stars)
• Requires a running MCP client to be useful
• Documentation and examples may be sparse

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.