Dbt Docs MCP

1[![Verified on MseeP](https://mseep.ai/badge.svg)](https://mseep.ai/app/ad4aaf73-63ce-42e0-b27c-8541ae1fbab8)
2 
3[![Trust Score](https://archestra.ai/mcp-catalog/api/badge/quality/mattijsdp/dbt-docs-mcp)](https://archestra.ai/mcp-catalog/mattijsdp__dbt-docs-mcp)
4 
5# dbt-docs-mcp
6 
7Model Context Protocol (MCP) server for interacting with dbt project metadata, including dbt Docs artifacts (`manifest.json`, `catalog.json`). This server exposes dbt graph information and allows querying node details, model/column lineage, and related metadata.
8 
9## Key Functionality
10 
11This server provides tools to:
12 
13*   **Search dbt Nodes:**
14    *   Find nodes (models, sources, tests, etc.) by name (`search_dbt_node_names`).
15    *   Locate nodes based on column names (`search_dbt_column_names`).
16    *   Search within the compiled SQL code of nodes (`search_dbt_sql_code`).
17*   **Inspect Nodes:**
18    *   Retrieve detailed attributes for any given node unique ID (`get_dbt_node_attributes`).
19*   **Explore Lineage:**
20    *   Find direct upstream dependencies (predecessors) of a node (`get_dbt_predecessors`).
21    *   Find direct downstream dependents (successors) of a node (`get_dbt_successors`).
22*   **Column-Level Lineage:**
23    *   Trace all upstream sources for a specific column in a model (`get_column_ancestors`).
24    *   Trace all downstream dependents of a specific column in a model (`get_column_descendants`).
25*   **Suggested extensions:**
26    *   Tool that allows executing SQL queries.
27    *   Tool that retrieves table/view/column metadata directly from the database.
28    *   Tool to search knowledge-base.
29 
30## Getting Started
31 
321.  **Prerequisites:** Ensure you have Python installed and [uv](https://docs.astral.sh/uv/)
332.  **Clone the repo:**
34    ```bash
35    git clone <repository-url>
36    cd dbt-docs-mcp
37    ```
383.  **Optional: parse dbt manifest for column-level lineage:**
39    - Setup the required Python environment, e.g.:
40    ```bash
41    uv sync
42    ```
43    - Use the provided script `scripts/create_manifest_cl.py` and simply provide the path to your dbt manifest, dbt catalog and the desired output paths for your schema and column lineage file:
44    ```bash
45    python scripts/create_manifest_cl.py --manifest-path PATH_TO_YOUR_MANIFEST_FILE --catalog-path PATH_TO_YOUR_CATALOG_FILE --schema-mapping-path DESIRED_OUTPUT_PATH_FOR_SCHEMA_MAPPING --manifest-cl-path DESIRED_OUTPUT_PATH_FOR_MANIFEST_CL
46    ```
47    - Depending on your dbt project size, creating column-lineage can take a while (hours)
484.  **Run the Server:**
49    - If your desired MCP client (Claude desktop, Cursor, etc.) supports mcp.json it would look as below:
50    ```json
51    {
52        "mcpServers": {
53            "DBT Docs MCP": {
54            "command": "uv",
55            "args": [
56                "run",
57                "--with",
58                "networkx,mcp[cli],rapidfuzz,dbt-core,python-decouple,sqlglot,tqdm",
59                "mcp",
60                "run",
61                "/Users/mattijs/repos/dbt-docs-mcp/src/mcp_server.py"
62            ],
63            "env": {
64                "MANIFEST_PATH": "/Users/mattijs/repos/dbt-docs-mcp/inputs/manifest.json",
65                "SCHEMA_MAPPING_PATH": "/Users/mattijs/repos/dbt-docs-mcp/outputs/schema_mapping.json",
66                "MANIFEST_CL_PATH": "/Users/mattijs/repos/dbt-docs-mcp/outputs/manifest_column_lineage.json"
67            }
68            }
69        }
70    }
71    ```
72