io.github.us-all/dbtdatabases

dbt

Verified · 10 days ago

dbt MCP — manifest/run_results/sources/catalog parsing + DQ result tables (BigQuery/Postgres)

Install

claude mcp add dbt -- npx -y @us-all/[email protected]

Requires: DBT_PROJECT_DIR, DBT_TARGET_DIR, DBT_RUN_HISTORY_DIR, DQ_BACKEND, DQ_RESULTS_TABLE, DQ_SCORE_TABLE, GOOGLE_APPLICATION_CREDENTIALS, BQ_PROJECT_ID, PG_CONNECTION_STRING, DBT_ALLOW_WRITE, DQ_SCHEMA, DQ_COL_RUN_AT, DQ_COL_CHECK_TYPE, DQ_COL_STATUS, DQ_COL_DATASET, DQ_COL_TABLE_NAME, DQ_COL_SEVERITY, DQ_COL_FAILURE_COUNT, DQ_COL_MESSAGE, DQ_COL_CHECK_NAME, DQ_COL_SCORE_DATE, DQ_COL_SCOPE, DQ_COL_TIER, DQ_TIER1_TARGET_PCT, DBT_SLA_CONFIG_PATH, DBT_TOOLS, DBT_DISABLE

Our take

A dbt-focused server for reading manifests, catalogs, source metadata, run results, and data quality result tables across BigQuery and Postgres. It is mainly for analytics engineers and teams that already run dbt and want structured project context inside MCP workflows. The scope is clear and practical, but with no visible adoption signals here it reads as a niche utility that likely needs hands-on validation before wider use.

reviewed by hand · 2026-07-28

Something wrong or dead here? Report it