AI Interfaces
Reasoning Checkpoint Rail
An ordered run rail that makes intent, evidence, tool planning, and answer readiness visible at a glance.
Installation
Copy and paste the code into your project.
Accessibility notes
The checkpoint sequence is an ordered list and each stage exposes its status as text rather than color alone.
Reasoning rail
Run checkpoints
- 1
Intent
doneScope resolved
- 2
Evidence
done4 sources grounded
- 3
Tool plan
active2 calls approved
- 4
Final answer
queuedWaiting on tool result
Preview accent
const checkpoints = [{ label: "Intent", status: "done" }, { label: "Evidence", status: "done" }, { label: "Tool plan", status: "active" }, { label: "Final answer", status: "queued" }];
export function ReasoningCheckpointRail() {
return <ol>{checkpoints.map((checkpoint, index) => <li key={checkpoint.label}>{index + 1}. {checkpoint.label} · {checkpoint.status}</li>)}</ol>;
}<ReasoningCheckpointRail />Related components
Approval
Publish launch prompt
AI Interfaces
Approval Action Panel
A review panel with assignee chips and paired approval actions for launch workflows.
Pipeline
AI Interfaces
Data Pipeline Card
A simple horizontal pipeline for data prep, agent workflows, and release steps.
Human handoff
Ready to escalate?
complete
Decision summary
Generated
Supporting evidence
4 sources attached
Open risks
1 needs review
Next owner
Support lead
AI Interfaces
Handoff Readiness Card
A pre-escalation checklist for packaging AI decisions, evidence, risks, and ownership for human review.