@langwatch/scenario
    Preparing search index...

    Interface EvaluationResult

    The result of one evaluator on a scenario run. Field names match the results.evaluations entries of the run finished event.

    interface EvaluationResult {
        cost?: { amount: number; currency: string };
        details?: string;
        evaluatorId: string;
        inputs?: Record<string, string>;
        label?: string;
        name: string;
        passed?: boolean;
        required: boolean;
        score?: number;
        status: "passed" | "failed" | "scored" | "skipped" | "error";
    }
    Index
    cost? details? evaluatorId inputs? label? name passed? required score? status
    cost?: { amount: number; currency: string }
    details?: string

    The reason: judge details, "no golden_sql on this scenario", "no run_sql call in the trace", "the mapping returned nothing", or the error.

    evaluatorId: string

    The saved evaluator id, or the evaluator type when run by type.

    inputs?: Record<string, string>

    The resolved input values, each cut to 2000 characters.

    label?: string
    name: string
    passed?: boolean
    required: boolean
    score?: number
    status: "passed" | "failed" | "scored" | "skipped" | "error"

    Settings
    Member Visibility

    On This Page
    Properties