Data transformation · 1 skill

dbt MCP and Claude skills

Read dbt job runs, failed models and tests, source freshness and model health, and post what broke in last night's run, where, and since when.

Connect dbt to your agent (MCP)

dbt MCP server, dbt Labs' official server, in two forms (docs). The remote server runs on the dbt platform at https://<your-dbt-host>/api/ai/v1/mcp/ (copy it from Account settings > Access URLs) and covers the Semantic Layer, Discovery, Admin API and SQL tools. Sign in with OAuth (beta as of October 2026), or send headers: claude mcp add --transport http dbt https://<your-dbt-host>/api/ai/v1/mcp/ --header "Authorization: Token <token>" --header "x-dbt-prod-environment-id: <id>"; token connections need the Developer permission set. The self-hosted server, uvx dbt-mcp, also works with dbt Core alone: it reads DBT_PROJECT_DIR and DBT_PATH, plus DBT_HOST, DBT_TOKEN and DBT_PROD_ENV_ID for platform features. Both work in Codex and Cursor, and dbt is in Claude's connector directory.
Admin and Discovery APIs with a service token. Create a service token with the Job Viewer and Metadata Only permission sets, and store it in DBT_TOKEN with DBT_ACCOUNT_ID and DBT_HOST.
dbt Core artifacts. In the project repo, the agent can run dbt source freshness or read target/run_results.json after your scheduled build, with no token at all.

For the self-hosted server, set DISABLE_DBT_CLI=true and keep DISABLE_SQL=true (its default) so the agent reads metadata without running models or SQL. A Job Viewer and Metadata Only token can't trigger or edit jobs.

1 dbt skill

dbt Cloud failed runs and stale sources

Every morning: dbt Cloud runs that failed since yesterday with the step and model that broke, plus sources past their freshness threshold, and a retry question.

Data teamsEvery day at 7:30Asks you

What agents do with dbt

On the dbt platform, the Admin API (/api/v2/accounts/{account_id}/runs/) lists job runs with a status (10 success, 20 error, 30 cancelled), timings and the job they belong to, and each run's artifacts include run_results.json, which holds every model and test's status, execution_time and error message. The Discovery API (GraphQL) answers questions across runs: model execution history, test results, source freshness and lineage. With dbt Core, the same files land in your project's target/ folder after dbt build, dbt test or dbt source freshness (which writes sources.json). The dbt MCP server wraps these in tools such as get_model_health, get_model_performance, get_lineage, list_metrics and query_metrics.

Data teams schedule a morning report on the overnight run (failed models and tests with the error and the owner), sources past their freshness threshold, the slowest models of the week, and a Semantic Layer metric posted as a number with its trend.

getdbt.com

Teams that use these

Your dbt numbers, on one board with everything else

Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.

Join the open beta Browse all skills

Free during the open beta. No card.