Starburst uses GateRook’s Trino / Presto connector. Starburst Galaxy and Starburst Enterprise through the Trino connector.
The screenshots show GateRook 2.0.0: the Starburst form, the endpoint selector, and this connector’s actual tool preview. The form and preview have not been saved. Select a screenshot to open it at full size.
1. Create a connection and enter the settings
In GateRook, open Connections, select Add connection, search for Starburst, choose its card and select Continue to configure.
Copy the Starburst cluster's connection host and port, and obtain a user that can read the required catalog and schema. Use the SQL coordinator endpoint rather than a management API.
For Starburst Galaxy, connection instructions can include a role in the username, such as email/role. Use the exact identity and role assigned for this connection.
Enter a Name, for example starburst-demo. Your AI app uses this name to choose the connection.
Enter the HTTPS endpoint in Coordinator URL and select Trino for Engine. Use the cluster's connection details, including any required port.
Choose User and password under Sign in with and enter User and Password. Put a Galaxy role-qualified username directly into User. Use JWT bearer token only when the cluster is configured to accept the supplied JWT.
Set Default catalog and Default schema, then restrict Catalogs to those agents need. Add CA certificate (PEM) only if the endpoint uses a private CA.
Choose Development, Staging or Production to match the service. Review the connector’s limit and access settings before continuing.
2. Test the connection and add it to an endpoint
Select Continue to test. GateRook checks the settings from your computer. Review the result and resolve any connection, certificate or permission error before continuing.
If login succeeds but a catalog is missing, check the selected Galaxy role and its catalog grants. If the endpoint responds with a website rather than SQL protocol responses, verify Coordinator URL against the cluster connection details.
After the test passes, select Continue to add to an endpoint. Choose the endpoint your AI app uses, review its tool modes, then select Save connection. GateRook offers default as the starter endpoint.
3. Use the connection from your AI IDE
After saving Starburst on the endpoint and passing its connection test, ask your AI app to use starburst-demo, or the name you chose. If its tools are missing, confirm the endpoint assignment and refresh the MCP connection or start a new chat.
The preview below shows this connector’s default tools before saving. The table explains what each tool can do; connection settings, credentials and endpoint policies can restrict access further.
Start with a small read request. Replace example resource names with existing resources your connection may access. Open Activity after the call to inspect its connection, tool, arguments, policy decision and result metadata.
Ask your AI, for example
- “List the catalogs and schemas I am permitted to read.”
- “Describe the columns of the orders table in the selected catalog.”
- “Return the order count by status for the last seven days, using a bounded read-only query.”
| Tool | What it does | Starts as |
|---|---|---|
trino_list_catalogs | Catalogs agents may use on the coordinator. | Read only |
trino_list_schemas | Schemas of one catalog. | Read only |
trino_list_tables | Tables and views of one schema, optionally matching a LIKE pattern. | Read only |
trino_describe_table | Columns with their types and comments, and the table's comment. | Read only |
trino_query | One SELECT, WITH, VALUES, SHOW, DESCRIBE or EXPLAIN. Trino's own IO plan checks it first: it must write no table and read only catalogs of the connection. Results end at the row limit: Trino can't resume a query, so narrow it to see more. | Read only |
trino_execute | One INSERT, UPDATE, DELETE, MERGE, CREATE, ALTER, DROP, COMMENT, TRUNCATE, ANALYZE or REFRESH statement. You approve the exact text with the tables Trino's plan reads and writes. | Approval |
You can change each tool’s mode on the endpoint. Every call your AI makes appears in Activity.