01Hash-based drift detection to identify stale or missing documentation
02Token-optimized markdown subsection extraction for lower context usage
03Automated documentation scraping from geminicli.com via llms.txt
04Natural language and keyword-based search for rapid doc discovery
05Resilient doc_id resolution and comprehensive index maintenance
061 GitHub stars