Type Alias CalibrateContentResponse

Type Declaration

• {
      verdict: BinaryVerdict;
      confidence?: number;
      explanation?: string;
      features?: {
          feature_id: string;
          status: FeatureCheckStatus;
          policy_id?: string;
          explanation?: string;
          confidence?: number;
      }[];
      context?: ContextObject;
      ext?: ExtensionObject;
  }

  • verdict: BinaryVerdict

  • Optionalconfidence?: number

    Model confidence in the verdict (0-1)

    Minimum

    0

    Maximum

    1

  • Optionalexplanation?: string

    Detailed natural language explanation of the decision

  • Optionalfeatures?: {
        feature_id: string;
        status: FeatureCheckStatus;
        policy_id?: string;
        explanation?: string;
        confidence?: number;
    }[]

    Per-feature breakdown with explanations. Mirrors validate_content_delivery feature shape so calibration loops can correlate against production verdicts by policy_id.

  • Optionalcontext?: ContextObject

  • Optionalext?: ExtensionObject

• { errors: Error[]; context?: ContextObject; ext?: ExtensionObject }