Before you start
You need:- Account Admin permissions in Sidecar
- A connected dbt Cloud integration
- An Access URL saved on the integration’s Connection tab, including for accounts that use standard URLs
- A production deployment environment accessible to the integration credential
- A dbt Cloud service account token with access to the projects and operations your agents need
Network access
Agent Console connects to Sidecar’s dbt Cloud MCP proxy. The proxy sends requests to your dbt Cloud Access URL and supplies the production environment header. Upstream requests originate from Sidecar’s production outbound IP,44.236.246.98/32.
If your dbt Cloud deployment restricts inbound access, allow Sidecar’s upstream
requests. Anthropic connects to the Sidecar proxy rather than directly to dbt
Cloud. For a custom direct endpoint, follow the two-client network
requirements.
Enable dbt Cloud MCP in Sidecar
- In Sidecar, open Integrations → dbt Cloud.
- On the Connection tab, confirm that your connection and Access URL are saved.
- Open the MCP tab.
- Click Enable MCP.
- Paste your token into Service token (organization-wide).
- Click Add server.
- In Integrations → MCP Servers, confirm that the server is connected.
Sidecar currently selects the first production environment it finds while
checking the account’s projects. It does not automatically switch to the
environment of each failing run. Keep this limitation in mind if your account
has multiple production environments.
Finish agent setup
Registering the server does not give it to an agent automatically. Grant the dbt Cloud MCP server to an agent and select it for jobs. For native dbt Firefighter jobs, grant the server to the Reliability agent. Sidecar adds the matching dbt Cloud server to the job’s connections when it is available. This optional connection works alongside the job’s warehouse connection. The connection exposes investigation tools and retries. It does not exposetrigger_job_run, cancel_job_run, execute_sql, or Semantic Layer tools.
To let an agent inspect or change dbt source code, also connect the
repository.
To rotate credentials or refresh tools, use the row actions in
Integrations → MCP Servers. See Manage a server.
Verify Agent Console access
After granting the server and selecting it for a job, start a new manual run. Ask the agent to use the dbt Cloud MCP server to inspect the status and error details of an existing job run the service token can access. Provide the run ID and project name or ID. Ask it to report the actual tool result without retrying or triggering a job. Expand the MCP call in the run transcript and confirm that it returns the intended job run’s details. A Connected badge alone does not establish runtime access. See runtime verification.Troubleshooting
No production environment could be resolved
No production environment could be resolved
Confirm that the account has a production deployment environment and that the
integration credential can access its projects and environments. Then save
the MCP credential again.
The Account ID or Access URL changed
The Account ID or Access URL changed
In Integrations → MCP Servers, select Replace token for the dbt Cloud
server and save its service token again. Sidecar must resolve the production
environment again before the connection can be used.
The agent cannot see dbt Cloud tools
The agent cannot see dbt Cloud tools
Confirm that the server is granted to the agent and enabled for the job,
then start a new run. Check the service token and Sidecar’s upstream network
access if initialization fails. A Connected badge can retain an earlier
authenticated result. For native dbt Firefighter jobs, the server must belong
to the dbt Cloud integration used by the job.