ScyllaDB uses GateRook’s Apache Cassandra connector. ScyllaDB and ScyllaDB Cloud through CQL.
The screenshots show GateRook 2.0.0: the ScyllaDB 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 ScyllaDB, choose its card and select Continue to configure. GateRook downloads the official Apache Cassandra connector module when you choose it. Wait for the download and verification to finish before configuring the connection.
Copy the CQL contact points, credentials and local data-center name from the ScyllaDB Cloud Connect view, or obtain the equivalent details from your self-managed cluster administrator.
Allow the GateRook computer through the cluster's network controls. The driver can discover additional nodes, so the reachable address of one seed node is not always sufficient.
Enter a Name, for example scylladb-demo. Your AI app uses this name to choose the connection.
Add the CQL endpoints to Contact points as host:port, using the supplied port. Set Local data center to the exact cluster value, rather than a guessed cloud region.
Enter the CQL User and Password. Choose TLS for Connection type when required, and add CA certificate (PEM) under Advanced for a private CA.
Set Keyspace for statements that omit a keyspace, then restrict Keyspaces to the application keyspaces agents need. Keep Allow ALLOW FILTERING off and use the required Read consistency.
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.
A local-data-center mismatch or unreachable discovered node can prevent session creation even when a contact point responds. For a rejected query, check its partition-key restrictions and the user's SELECT permission rather than enabling a full-cluster scan.
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 ScyllaDB on the endpoint and passing its connection test, ask your AI app to use scylladb-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 permitted keyspaces and tables.”
- “Describe the primary key and columns of the orders table, if present.”
- “Read at most 10 rows from orders for one known partition key; do not use ALLOW FILTERING.”
| Tool | What it does | Starts as |
|---|---|---|
cassandra_list_keyspaces | Keyspaces agents may use, with their replication, tables and materialized views, from system_schema. | Read only |
cassandra_describe_table | Columns of one table or view with their kinds (partition key, clustering, static, regular), types and clustering order, its secondary indexes and views, default TTL and compaction. | Read only |
cassandra_query | One CQL SELECT with ? or :name values, at the connection's read consistency. Pages with nextCursor. ALLOW FILTERING scans the cluster and is refused unless a dev connection allows it. | Read only |
cassandra_execute | One INSERT, UPDATE, DELETE or TRUNCATE, or CREATE, ALTER or DROP of a table, index, type or materialized view. You approve the exact statement and its values; it runs once, never retried. | Approval |
You can change each tool’s mode on the endpoint. Every call your AI makes appears in Activity.