Risk Cards With Icons

The "what could go wrong" panel — risks made scannable through visual coding.

What this device does

Replaces a "Risks" bulleted list with a card grid. Each card carries an emoji (signals the risk domain at a glance), a severity-colored left border (red/orange/yellow for high/med/low), a short title, and one sentence. Reader can absorb 4 risks in under 15 seconds.

When to use

• "What could go wrong" panels in vivid recipes
• Always pair with prose committing to a kill-switch ("we'd cut the position if any escalates")
• NOT for: institutional reports (those want full risk paragraphs with mitigants)

How agents use this

data = {
  risks: [
    { icon: "🇨🇳", title: "Greater China", severity: "high", text: "18% of revenue, -2% YoY. Stabilized for now but local competition + geopolitics could re-accelerate weakness." },
    { icon: "⚖️", title: "App Store regulation", severity: "med", text: "EU DMA enforcement live. US DoJ remedies pending. A 10pp take-rate cut is a real downside scenario." },
    { icon: "🤖", title: "AI execution", severity: "med", text: "If Apple Intelligence underwhelms in iOS 19, the replacement-cycle thesis weakens fast." },
    { icon: "📊", title: "Factor reversal", severity: "low", text: "AAPL screens long Quality, short Momentum. A sudden value rotation hurts (but doesn't break) the call." }
  ]
}

Visual register

2-column grid (1-column on mobile). Each card has 4px left border in severity color (red #ff5577 high, orange #ff8c42 med, yellow #ffd166 low). Icon at 24px above title.

Example output

See ./preview.html — AAPL Buy thesis risks, April 2026.