Interface TrustVerdict

The full plain-language verdict for a model, derived from its metrics.

interface TrustVerdict {
    canAct: boolean;
    explanation: string;
    gateReason: string;
    grade: TrustGrade;
    headlineMetric: { key: "AUC" | "Accuracy" | "R2"; value: number };
    oneLiner: string;
    score01: number;
    unknown: boolean;
}

Properties

canAct: boolean

Whether the user may ACT on this prediction (the gate). false for Poor or unmeasured models.

explanation: string

Guidance for how much to rely on it (grade-specific).

gateReason: string

When canAct is false, the plain reason actions are locked; otherwise null.

The plain reliability grade.

headlineMetric: { key: "AUC" | "Accuracy" | "R2"; value: number }

The metric that drove the grade, for the "For analysts ▸" disclosure. null if none usable.

oneLiner: string

Headline in plain words, e.g. "Right about 8 out of 10 times."

score01: number

0–1 normalized trust score, for a meter/visual (NOT a probability).

unknown: boolean

True when no usable quality metric was found — trust is unknown and actions are gated off.