dbt Cloud MCP server

What the model can do

dbt Cloud: projects, environments, scheduled jobs with their last run, job runs by job or status (failed, running, succeeded) with duration and branch, and one run's steps and, when asked, their logs. Can trigger a job run, retry or re-run a failed one, cancel a run, or delete a job, when asked.

Reads

• dbtcloud_list_projectsDbt Cloud projects in the account: name, description, repository and subdirectory.
• dbtcloud_list_environmentsThe environments of a dbt Cloud project (development, staging, production): name, type and deployment type.
• dbtcloud_list_jobsDbt Cloud jobs: name, schedule, environment, the next scheduled run, and the status and time of the most recent run and the most recent finished run.
• dbtcloud_list_runsDbt Cloud job runs, newest first: job, status (queued, running, success, error, cancelled), message, when created and finished, duration…
• dbtcloud_get_runOne dbt Cloud run with each step (dbt deps, dbt build…) and its status and duration.

Acts, when you ask

These change something in dbt Cloud. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• dbtcloud_trigger_job_runStarts a run of a dbt Cloud job now.
• dbtcloud_cancel_runCancels a queued or running dbt Cloud run.
• dbtcloud_retry_runRetries a failed dbt Cloud run from its point of failure, as a new run.
• dbtcloud_rerun_jobRe-runs a dbt Cloud job: from the point of failure if its last run failed, otherwise as a fresh full run.
• dbtcloud_delete_jobPermanently deletes a dbt Cloud job definition: its schedule stops for good and it cannot be restored.

Connect it

First connect dbt Cloud on Dashboard → Connections, then add Memtro to your client once. The dbt Cloud tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared dbt Cloud connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Did last night's dbt build pass?
• › Which dbt runs failed today, and why?
• › Show me the steps and logs of the last failed nightly run
• › When is the production job next scheduled to run?
• › Retry last night's failed production run from where it stopped

Questions

Is this an official dbt Cloud MCP server?

No. It is Memtro's MCP server with Memtro's dbt Cloud connector behind it: the connector reads dbt Cloud through its API with the credentials you connect on Dashboard → Connections, and dbt Cloud publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in dbt Cloud?

Only when you ask, and only what it has been granted: trigger a run of a dbt Cloud job now (runs dbt against the warehouse); cancel a queued or running dbt Cloud run; retry a failed dbt Cloud run from its point of failure; re-run a job: from the failure point if its last run failed, otherwise afresh; delete a dbt Cloud job and its schedule (permanent). Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Databases and warehouses

• Postgres MCP server
• MySQL and MariaDB MCP server
• SQL Server MCP server
• BigQuery MCP server
• Snowflake MCP server
• AWS Athena MCP server