Skip to main content
POST
Create an agent outcome rubric

Authorizations

Authorization
string
header
required

Dashboard JWT token from Clerk

Path Parameters

agentId
string
required

Body

application/json
name
string
required

Display name for the rubric.

Required string length: 1 - 200
criteria_md
string
required

Markdown success criteria the LLM judge grades against.

Required string length: 1 - 10000
evaluator_prompt
string

Optional custom judge system prompt. Omit (or pass "default") to use the built-in prompt.

Maximum string length: 8000
scope
enum<string>

Which conversations the rubric applies to. Defaults to agent.

Available options:
agent,
inbox,
voice,
org

Response

201 - application/json

The rubric was created.

data
object
meta
object