Normalize data

Sources do not share a format. RSS, HTML list pages, and news APIs all become the same item fields so search, alerts, and workflows do not care which connector produced a row.

Why it matters

RSS ───────┐
HTML list ─┤
News API ──┘
      ↓
  Normalized item
      ↓
Search · Feeds · Alerts · Workflows

What downstream uses

• Full-text search over title, summary, and body
• Category streams on Feeds and the Feed API
• Alert match on keyword or category
• Workflow conditions on category, title, summary, URL, source_id

Normalization is the foundation of the rest of the platform. Intelligence products (PolyIntel, PolyRAG) will sit on the same items when they ship.

• Search
• Indexing
• Watches and monitoring
• Feeds