Agent Settings
Create Evals Bulk
Create multiple evals for an agent in a single request.
Equivalent to calling POST /settings/evals once per item, but atomic
from the caller’s perspective.
POST
Create Evals Bulk
Authorizations
Path Parameters
Body
application/json
Extra context the AI sees when scoring a call; the transcript is always included. Allowed values: system_prompt, tools, allowed_capabilities, workflow (the agent's configuration), or metadata.<key> for a value from the call's metadata, with dots for nested keys (e.g. metadata.account.region). Keys are case-sensitive, and a key a call does not have is skipped for that call. Any other value returns a 422. Defaults to none.
Examples:
Response
Successful Response