Skip to main content
A connector points the assistant at something of yours: a Postgres or MySQL database, a server over SSH, or an MCP server. Credentials are sealed and only decrypted for the length of one call. Add one from the + menu in the composer.

The three kinds

Database. Postgres or MySQL. Host, port, database, username, password, and TLS settings if you need them. The assistant gets a run_sql_query tool. SSH. Host, port, username, and either a password or a private key with an optional passphrase. The assistant gets an ssh_run_command tool. MCP. A URL and a transport, either HTTP or SSE. Authentication can be none, a header, a bearer token or OAuth. The server’s own tools appear to the assistant with their real names and descriptions. New connectors are read only by default.

Nothing writes without you

Reads run. Anything that could change something stops and asks. For SQL, statements beginning select, with, show, explain, describe, values or table are reads. Everything else is a write and is gated. So is anything with multiple statements, and a CTE that hides an insert or delete inside it. Every SSH command is gated, with no read and write distinction, because a shell command is too easy to disguise. MCP tools skip the gate only when the server itself declares them read only.

Approving

When something is gated you get four choices. Commands judged dangerous, such as drop, truncate, grant, rm, dd, sudo or anything carrying shell metacharacters, are marked Dangerous action, review carefully and cannot be added to an allow list. They can only ever be approved one at a time. An allow list entry is scoped to one chat and one connector. It does not carry over to another conversation.
There is currently no screen for reviewing or revoking allow list entries. They live with the chat, so clearing the chat is the way to drop them.

Limits

Queries return at most 1,000 rows and are cancelled after 30 seconds. SSH commands are cancelled after 30 seconds and their output is truncated at 100,000 characters. There is no cap on how many connectors you may create and no plan gating on them.

MCP over OAuth

Choosing OAuth opens the provider in a new tab. Approve there, come back, and press Verify connection. Tokens refresh a minute before they expire. Disconnecting deletes the stored authorization outright.