detect-breaking-news
Identify articles that qualify as breaking news requiring immediate alerts. Fast check based on article metadata without additional LLM calls.
When & Why to Use This Skill
The Detect Breaking News skill is a high-efficiency diagnostic tool designed to identify critical, time-sensitive articles from a processed news feed. By leveraging metadata such as importance scores and breaking news flags, it enables real-time alerting systems to prioritize high-impact events—such as major product launches, security vulnerabilities, or regulatory shifts—without the latency or cost of additional LLM calls.
Use Cases
- Real-time AI Industry Monitoring: Automatically flag major model releases (e.g., GPT-5, Claude 4) or significant research breakthroughs for immediate executive briefing.
- Automated Newsroom Alerts: Enable digital publishers to instantly identify and push notifications for high-importance stories while filtering out routine updates.
- Crisis & Security Tracking: Monitor tech news to isolate critical security vulnerabilities or safety incidents that require immediate attention from SRE or security teams.
- Market Intelligence: Track major company announcements, acquisitions, or leadership changes to provide traders or analysts with a competitive edge.
| name | detect-breaking-news |
|---|---|
| description | Number of breaking news articles detected |
| domain | news |
| category | diagnostic |
| mcp-servers | [] |
| requires-approval | false |
| confidence | 0.95 |
| - name | breaking_count |
| type | int |
Detect Breaking News
Identify articles that require immediate breaking news alerts.
Preconditions
- Articles have been processed with analyze-article skill
- Articles have
is_breakingflag andimportance_scoreset
Actions
Step 1: Apply Breaking News Criteria
For each article, check if it meets ALL criteria:
- is_breaking: Must be true (set by LLM during analysis)
- importance_score: Must be >= 8 (high or critical importance)
Both conditions must be met to prevent:
- Low-importance articles flagged as breaking (score < 8)
- High-importance routine updates (is_breaking = false)
Step 2: Collect Breaking Articles
Add all qualifying articles to the breaking news list.
Step 3: Return Results
Return list of breaking articles and count.
Success Criteria
- All articles checked against criteria
- Breaking articles correctly identified
- No false positives (low importance flagged as breaking)
- No false negatives (true breaking news missed)
Example Breaking News Categories
Typical breaking news includes:
- Major model releases (GPT-5, Claude 4, Gemini 2, etc.)
- Security vulnerabilities in AI systems
- Major company announcements (acquisitions, leadership changes)
- Regulatory decisions affecting AI industry
- Significant research breakthroughs
- Safety incidents or model failures