Briefings

AI-generated intelligence summaries from your collected articles.

How Briefings Work

1. Collection - Recent articles gathered from your sources
2. Filtering - Verified articles prioritized (confidence > 70%)
3. Synthesis - LLM analyzes and summarizes key developments
4. Formatting - Structured output with sources and confidence scores

Generating Briefings

Manual Generation

Click Generate Briefing in the dashboard, or:

curl -X POST http://localhost:3000/api/briefings/generate \
  -H "Content-Type: application/json" \
  -d '{"hours": 24}'

Scheduled Generation

Configure automatic briefings:

# Generate briefing every day at 5am and 6pm
BRIEFING_SCHEDULE="0 5,18 * * *"

Briefing Options

Parameter	Description	Default
hours	Time window to analyze	24
domains	Filter by domains	All
minConfidence	Minimum verification score	0.7
maxArticles	Cap on articles to include	50

Output Format

## Strategic Intelligence Briefing
**Generated:** 2026-02-07 05:00 EST
**Period:** Last 24 hours
**Sources:** 15 feeds | **Articles:** 42 analyzed

### 🔥 Priority Developments

1. **[Title]**
   [Summary of key points]
   Sources: Reuters, AP, BBC
   Confidence: 94%

### 📊 Domain Summary
- Geopolitics: 12 articles, 3 high-priority
- Economics: 8 articles, 2 high-priority
- Technology: 15 articles, 4 high-priority

### 📈 Trends
- Increasing coverage of [topic]
- New developments in [area]

Delivery

Briefings can be delivered via:

• Dashboard - View in web interface
• Telegram - Push to configured chat
• API - Fetch programmatically
• Webhook - POST to custom endpoint