Start from a root URL, follow the paths that matter, and keep ingestion scope aligned with the content your AI systems actually need.
Process sites that rely on frontend rendering, dynamic navigation, or client-side content assembly without rewriting custom scrapers for each target.
Prepare clean chunks with metadata so downstream retrieval, search, and grounding workflows do not start from raw website HTML.
Move from crawl to delivery in a controlled workflow once the workspace is active and connector settings are configured in the dashboard.