Data & AI · wired today
What can agents actually do with Fivetran?
Models and datasets are infrastructure now, and agents need eyes on them. The Fivetran connector is one of 205 executable connectors and gives agents 4 read-only tools, with the credential held in the vault and every call receipted.
The Fivetran tool list, exactly as agents see it
list_groups— List Fivetran groups (destination workspaces) in the account, newest cursor page firstlist_connectors— List the connectors inside one Fivetran group, with sync and setup state per connectorget_connector— Get one Fivetran connector by ID, including its setup and sync statuslist_destinations— List the destinations (warehouses) configured in the Fivetran account
Every tool above is read-only. Write surfaces ship only after live testing, and always behind approval gates.
What the Fivetran connector needs
You provide API key, API secret from your Fivetran account; Quox sends it as HTTP basic credentials. The credential is stored encrypted in the vault and resolved just-in-time, so agents use it without ever seeing it.
Where to get it: Fivetran's own API documentation ↗ covers creating and scoping the credential.
How proven is this?
This connector was blind shape-proven against the real Fivetran API: every endpoint it declares answered measurably differently from a deliberately bogus control request, without any account or credential involved. It is Beta: proven against the live API surface, not yet run with a real credential.
Fivetran integration FAQ
What can Quox agents do with Fivetran?
Agents can run 4 read-only tools against Fivetran: list groups, list connectors, get connector, list destinations. Every call is receipted in the evidence trail.
What does the Fivetran integration need?
You provide API key, API secret from your Fivetran account; Quox sends it as HTTP basic credentials. The credential is stored encrypted in the vault and resolved just-in-time, so agents use it without ever seeing it.
Do agents see my Fivetran credentials?
No. Credentials live encrypted in the Quox vault and are resolved server-side at call time. Agents invoke tools; they never receive the underlying secret.
How proven is the Fivetran connector?
This connector was blind shape-proven against the real Fivetran API: every endpoint it declares answered measurably differently from a deliberately bogus control request, without any account or credential involved. It is Beta: proven against the live API surface, not yet run with a real credential.