Skip to main content
GET
Retrieve an Eval template

Authorizations

X-Lightsage-Api-Key
string
header
required

Path Parameters

template_id
string
required

Response

Successful Response

name
string
required

Organization-visible name for this reusable template.

Required string length: 1 - 160
included_components
enum<string>[]
required

Settings copied when the template is applied. Omitted components do not override the new eval; included components copy their saved value, including an empty value.

Minimum array length: 1
Available options:
judges,
targets,
persona,
competitors,
tag,
category,
runs,
schedule,
repository,
environment,
search_grounding,
skills,
mcp_servers,
clis,
brand_docs,
github_pr,
runtime_commands
id
string
required
org_id
string
required
version
integer
required
created_at
required
updated_at
required
description
string | null

Optional guidance describing when to use the template.

Maximum string length: 1000
prompt
string | null

Prompt filled into the new eval form when the template is applied, or the scenario for Agent Preference, where {{competitor_name}} names the brand each comparison is against. Requests that create evals from a template still supply their own prompts.

Maximum string length: 20000
eval_type
enum<string>
default:custom_eval

Kind of eval the template creates. A template only applies to evals of the same type.

Available options:
custom_eval,
agent_preference
persona
string | null

Optional persona copied when persona is included.

Maximum string length: 500
category_name
string | null

Optional category copied when category is included.

Maximum string length: 160
tag
string | null

Optional tag copied when tag is included.

Maximum string length: 100
competitor_ids
string[]

Competitor IDs that create additional variants when competitors is included.

judges
GraderDefinition · object[]

Judge definitions copied when judges is included. May be empty to explicitly create the eval without template judges.

Maximum array length: 20
target_keys
string[]

Runnable target keys copied when targets is included, such as agent:codex:gpt-5.6-terra.

schedule
EvalTemplateSchedule · object

Schedule copied when schedule is included.

runs_per_execution
integer
default:1

Attempt count copied when runs is included.

Required range: 1 <= x <= 20
runtime
EvalTemplateRuntime · object

Saved runtime values. Each value is copied only when its matching component appears in included_components.

created_by
string | null
deleted_at