Skip to main content
DBDock Cloud exposes a hosted Model Context Protocol (MCP) server, so AI assistants can operate your DBDock account on your behalf — safely, with your permission, over secure OAuth sign-in.

​
Endpoint

The server uses the streamable HTTP transport and browser-based OAuth sign-in. There are no API keys to create or paste — you approve access in your browser, the same way you sign in to the dashboard.

​
Connect a client

Or add it to your MCP config:
The first time a client connects, your browser opens to authorize it. After that, the client acts within your DBDock account.

​
What the assistant can do

Once connected, the assistant can call DBDock tools such as:

list_connections

See the databases in your account.

create_backup

Start a backup of a connection.

restore_backup

Restore a backup into a database.

copy_database

Clone one database into another.

run_migration

Run a cross-database migration.

list_schedules · run_schedule_now · set_schedule_enabled

Inspect and control schedules.
MCP tools act on real databases. Backups, restores, and copies through the assistant have the same effect as running them in the dashboard — review what an assistant proposes before you approve it, especially restores and copies, which overwrite data.

​
Security

The MCP server authenticates every request through DBDock’s OAuth flow and scopes access to your own account. It never exposes your database credentials to the client — the assistant asks DBDock to perform actions; it doesn’t receive your secrets. See Cloud security.