Skip to main content
Connect your Databricks workspace to let CloudThinker agents review clusters, jobs, runs, notebooks, files, and catalog assets, run focused data queries, and diagnose failed work before anyone touches production. Databricks authenticates with a personal access token for your workspace. Operational changes stay deliberate, approval-based decisions.

​
Prerequisites

  • A Databricks workspace and a user allowed to create personal access tokens in it.
  • Optionally, a SQL warehouse if you want agents to run SQL queries.

​
Setup

1

Create a personal access token

In your Databricks workspace, click your username in the top bar and select Settings → Developer. Next to Access tokens, click Manage, then Generate new token. Set a lifetime and copy the token.
2

Find your workspace URL and warehouse ID

The workspace URL is your Databricks address, such as https://dbc-a1b2345c-d6e7.cloud.databricks.com. For SQL queries, copy the warehouse’s ID from SQL Warehouses → your warehouse → Connection details — it is the last segment of the HTTP path.
3

Add the connection in CloudThinker

Navigate to Connections → Databricks and enter:
  • DATABRICKS_HOST: your workspace URL
  • DATABRICKS_TOKEN: the personal access token
  • DATABRICKS_WAREHOUSE_ID: optional — only needed for SQL queries
Click Connect. CloudThinker verifies the credentials and shows a Connected status.
Databricks automatically revokes personal access tokens that have not been used for 90 days, and a token expires at the lifetime you set. Recreate the token and update the connection when either happens.

​
Connection details

​
Required permissions

The connection acts as the user who created the token, so scope that user deliberately.
Create the token as a user with read access to the clusters, jobs, and catalog assets agents should investigate — not a workspace admin. Leave DATABRICKS_WAREHOUSE_ID empty if you do not want agents running SQL.

​
Agent capabilities

Once connected, agents can:

​
Verify the connection

​
Example prompts

​
Troubleshooting

The token is wrong, expired, or was revoked — Databricks revokes tokens unused for 90 days. Generate a new token and update the connection.
DATABRICKS_HOST is wrong or unreachable. Use the full workspace URL, including https://, exactly as it appears in your browser.
DATABRICKS_WAREHOUSE_ID is missing or wrong, or the warehouse is stopped. Copy the ID from the warehouse’s Connection details and confirm the warehouse can start.

​
Security

  • Least privilege — grant only the permissions the agents need for your use case; start read-only and widen later.
  • Read-only by default — use read-only credentials unless you want agents to make changes through this connection.
  • Rotate credentials — rotate keys and tokens on your normal schedule; CloudThinker picks up the new value when you update the connection.
  • Revoke on offboarding — remove the credential at the provider when you delete a connection or a teammate leaves.
  • Token identity — the connection can do whatever the token’s user can do, so create it from a deliberately scoped user.
  • Opt in to SQL — leaving the warehouse ID empty keeps the connection to cluster, job, and catalog review.

Snowflake Connection

Warehouse cost and query analysis

Tony Agent

Database analysis and optimization