@langwatch/scenario
    Preparing search index...

    Interface RunEvaluatorsDeps

    What the runner needs from LangWatch and from the run.

    interface RunEvaluatorsDeps {
        evaluate: (
            args: {
                data: Record<string, unknown>;
                evaluatorRef: string;
                settings?: Record<string, unknown>;
                traceId?: string;
            },
        ) => Promise<EvaluateApiResponse>;
        fetchRemoteTraces?: () => Promise<void>;
        getEvaluatorSpec: (
            evaluatorRef: string,
        ) => Promise<EvaluatorSpec | undefined>;
    }
    Index
    evaluate fetchRemoteTraces? getEvaluatorSpec
    evaluate: (
        args: {
            data: Record<string, unknown>;
            evaluatorRef: string;
            settings?: Record<string, unknown>;
            traceId?: string;
        },
    ) => Promise<EvaluateApiResponse>
    fetchRemoteTraces?: () => Promise<void>

    Fetches the remote traces of the run into the span collector, waiting the configured budget. Called at most once, only when a mapping read the trace and found nothing while the messages carry trace ids.

    getEvaluatorSpec: (evaluatorRef: string) => Promise<EvaluatorSpec | undefined>

    Settings
    Member Visibility

    On This Page
    Properties