Skip to main content

PromptEvaluatorsClient

adaline.prompts.evaluators manages the evaluators attached to a prompt — LLM-as-a-judge graders, JavaScript checks, text matchers, cost, latency, and response-length guards. Evaluators are always scoped to a prompt; there is no workspace-level evaluators collection.

Access

The class is also exported directly:
Types from @adaline/api:
Evaluator types at a glance:

list()

List evaluators attached to a prompt (paginated).

Parameters

Returns

Promise<ListEvaluatorsResponse> with { data: Evaluator[]; pagination: Pagination }.

Example


create()

Attach a new evaluator to a prompt.

Parameters

Returns

Promise<Evaluator> — the newly attached evaluator with its server-assigned id.

Example — LLM-as-a-judge

Example — text matcher


get()

Fetch a single evaluator by ID.

update()

Update an evaluator’s title, settings, or threshold.

Example


delete()

Permanently delete an evaluator. Results from past evaluations that used it are preserved.

See Also