AI Safety endpoints
Evaluate
Evaluation and guardrails
POST
/
v1
/
evaluate
Authorizations
Authorization
string
headerrequiredBearer authentication header of the form Bearer <token>
, where <token>
is your auth token.
Body
application/json
messages
object[]
requiredRails and policies will be evaluated based on the messages. Policies requires at least the user and assistant, and system message is optional.
application
string
default: your_application_idpolicy_ids
string[]
If provided, will override the default policies enabled on console.
policies
object[]
override_response
string | null
correction_enabled
boolean
default: falseIf enabled and any violation happens, will provide a compliant correction to the user input. Override_response has priority.
fail_fast
boolean
default: trueResponse
200 - application/json
time
number
requiredcreated
integer
requiredstatus
string
requiredobject
string
default: evalpolicy_violations
object | null
correction
object | null