Langfuse JS/TS SDKs
    Preparing search index...

    Interface EvaluationRule

    Live evaluation rule for incoming observations.

    A rule determines which evaluators should be used, which observations should trigger scoring, how often scoring should run, and which observation fields should populate prompt variables.

    interface EvaluationRule {
        createdAt: string;
        createdBy: null | Creator;
        enabled: boolean;
        evaluatorAssignments: EvaluatorAssignment[];
        filter: EvaluationRuleReadFilter[];
        id: string;
        name: string;
        sampling: number;
        updatedAt: string;
    }
    Index

    Properties

    createdAt: string

    Timestamp when the evaluation rule was created.

    createdBy: null | Creator

    User who created this rule, or null when no user can be resolved.

    enabled: boolean

    Whether live execution is enabled for this rule.

    evaluatorAssignments: EvaluatorAssignment[]

    Evaluators attached to this rule in deterministic assignment order.

    List of stored filter conditions returned verbatim. Filters with a key use the keyed response shape; all others use the base shape. These response shapes are not broken down by internal filter type. An empty list matches every incoming object.

    id: string

    Stable evaluation-rule identifier.

    name: string

    Human-readable rule name. This is independent from evaluator names and does not need to be unique.

    sampling: number

    Fraction of matching observations that should be evaluated.

    Must be between 0 and 1.

    • 1 evaluates every matching observation.
    • 0.25 evaluates approximately 25% of matching observations.
    updatedAt: string

    Timestamp when the evaluation rule was last updated.