Skip to main content
GET
Get annotation task

Authorizations

X-API-Key
string
header
required

API key for authentication

Path Parameters

task_uuid
string
required

Task to retrieve

Example:

"f47ac10b-58cc-4372-a567-0e02b2c3d479"

Response

Successful Response

uuid
string
required

Unique ID for the task

Required string length: 36
Example:

"f47ac10b-58cc-4372-a567-0e02b2c3d479"

name
string
required

Name of the task

type
enum<string>
required

Task type. Determines the shape of each item's payload.

  • stt: judge a transcript on its own
  • llm: judge one response with its conversation history
  • llm-general: judge a standalone input -> output pair
  • conversation: judge a full conversation
Available options:
stt,
llm,
llm-general,
conversation,
tts
created_at
string
required

When the task was created (ISO 8601 UTC)

updated_at
string
required

When the task was last updated (ISO 8601 UTC)

description
string | null

The task's description

evaluators
object[]

The evaluators linked to this task, in display order. Fetching one task by ID enriches each with its live version's rubric. Listing tasks returns lightweight metadata (uuid, name, type, position) only

item_count
integer | null

Number of items in the task

has_agreement
boolean
default:false

Whether the task has at least one comparable human-vs-human or human-vs-evaluator pair, computed over all time

items
object[]

The task's items, each with its agreement stats. You get these when you fetch one task by ID, not when you list tasks

jobs
object[]

The task's labelling jobs. You get these when you fetch one task by ID, not when you list tasks