ReviewRequestCard
Card surface for human review requests from an agent. Shows the decision context, confidence, urgency level, and approve/reject actions.
import { ReviewRequestCard } from "@danagraph/design-system/patterns"Props
| Prop | Type | Default | Description |
|---|---|---|---|
requestIdreq | string | — | Unique review request identifier. |
agentNamereq | string | — | Agent requesting review. |
descriptionreq | string | — | What the agent wants reviewed. |
confidence | number | — | Agent confidence at time of request (0–1). |
urgency | "low" | "medium" | "high" | "critical" | — | Request urgency level. |
onApprove | () => void | — | Called on approval. |
onReject | (reason?: string) => void | — | Called on rejection. |
requestIdrequiredstringDefault —Unique review request identifier.
agentNamerequiredstringDefault —Agent requesting review.
descriptionrequiredstringDefault —What the agent wants reviewed.
confidencenumberDefault —Agent confidence at time of request (0–1).
urgency"low" | "medium" | "high" | "critical"Default —Request urgency level.
onApprove() => voidDefault —Called on approval.
onReject(reason?: string) => voidDefault —Called on rejection.
This component was designed to express:
Explore Related
Queue of pending human approvals with priority ordering
Structured justification form for human overrides of agent decisions. Captures reason category, free-text rationale, and acknowledges accountability.
High-stakes override with consequence preview and mandatory acknowledgment — no blind overrides, consequences always shown first
Traceable action attribution receipt. Shows decision authority, executing agent, human approver, confidence at decision time, and evidence used. Mandatory for consequential agentic actions.