Skip to content
sgp
API Reference

Evaluation Groups

Create Evaluation Group
POST/v5/evaluation-groups
List Evaluation Groups
GET/v5/evaluation-groups
Get Evaluation Group
GET/v5/evaluation-groups/{group_id}
Patch Evaluation Group
PATCH/v5/evaluation-groups/{group_id}
Update Evaluation Group
PUT/v5/evaluation-groups/{group_id}
Archive Evaluation Group
DELETE/v5/evaluation-groups/{group_id}
Get Evaluation Group Schema
GET/v5/evaluation-groups/{group_id}/schema
ModelsExpand Collapse
EvaluationGroup object { id, account_id, created_at, 8 more }

Response model for evaluation group

id: string

Unique identifier of the evaluation group

account_id: string

Account that owns this evaluation group

created_at: string

When the group was created

formatdate-time
name: string

Name of the evaluation group

tags: array of string

The tags associated with the entity

deleted_at: optional string

When the group was soft-deleted

formatdate-time
description: optional string

Optional description

members: optional array of EvaluationGroupMember { id, created_at, evaluation_group_id, 6 more }

Evaluation members in this group. Returned by default on GET by id; on list endpoints, populated only with the ‘members’ view.

id: string

Unique identifier of the member record

created_at: string

When this member was added to the group

formatdate-time
evaluation_group_id: string

ID of the evaluation group

evaluation_id: string

ID of the evaluation

deleted_at: optional string

When this membership was soft-deleted (if applicable)

formatdate-time
evaluation_created_at: optional string

When the evaluation was created

formatdate-time
evaluation_name: optional string

Name of the evaluation

evaluation_tags: optional array of string

Tags of the evaluation

object: optional "evaluation_group.member"
metadata: optional map[unknown]

Optional metadata key-value pairs

object: optional "evaluation_group"
row_identifiers: optional array of EvaluationGroupRowIdentifier { column_name, evaluation_id, object }

Row identifier mappings. Populated with ‘row_identifiers’ view.

column_name: string

Name of the column used as row identifier

evaluation_id: string

ID of the evaluation

object: optional "evaluation_group.row_identifier"
EvaluationGroupMember object { id, created_at, evaluation_group_id, 6 more }

Response model for evaluation group member

id: string

Unique identifier of the member record

created_at: string

When this member was added to the group

formatdate-time
evaluation_group_id: string

ID of the evaluation group

evaluation_id: string

ID of the evaluation

deleted_at: optional string

When this membership was soft-deleted (if applicable)

formatdate-time
evaluation_created_at: optional string

When the evaluation was created

formatdate-time
evaluation_name: optional string

Name of the evaluation

evaluation_tags: optional array of string

Tags of the evaluation

object: optional "evaluation_group.member"
EvaluationGroupRowIdentifier object { column_name, evaluation_id, object }

Response model for evaluation group row identifier

column_name: string

Name of the column used as row identifier

evaluation_id: string

ID of the evaluation

object: optional "evaluation_group.row_identifier"
EvaluationGroupViews = "members" or "row_identifiers"

Views for evaluation group responses

One of the following:
"members"
"row_identifiers"
EvaluationGroupRetrieveSchemaResponse object { evaluation_group_id, evaluation_schemas, object }

Per-evaluation schemas for all members of an evaluation group

evaluation_group_id: string

The ID of the evaluation group

evaluation_schemas: array of EvaluationSchemaResponse { evaluation_id, fields, total_items, 3 more }

Schema for each member evaluation in the group, one entry per active evaluation

evaluation_id: string

The ID of the evaluation

fields: array of object { data_type, field_name, item_count, 2 more }

List of all discovered fields, ordered alphabetically by field_name

data_type: string

JSON type: ‘string’, ‘number’, ‘boolean’, ‘object’, ‘array’, or ‘null’

field_name: string

The flattened JSON key path (e.g., ‘metadata.category’)

item_count: number

Number of evaluation items containing this field

minimum0
source: "data" or "task_result_cache"

The source of the field: ‘data’ or ‘task_result_cache’

One of the following:
"data"
"task_result_cache"
object: optional "field_schema"
total_items: number

Total number of evaluation items

minimum0
is_sampled: optional boolean

Whether schema was computed from a sample of items (for large evaluations)

object: optional "evaluation_schema"
sample_size: optional number

Number of items sampled for schema inference, if applicable

minimum0
object: optional "evaluation_group_schema"