📋 Feature Request

Add a document browser component that allows users to browse and filter document chunks from crawled websites with advanced domain filtering capabilities.

🎯 Problem

Users need a way to:

• Browse through document chunks from crawled sources
• Filter content by specific domains
• Search within document content
• View chunk metadata and details
• Navigate through large documents efficiently

✨ Proposed Solution

Document Browser Modal

• Two-column layout for chunk list and content viewing
• Domain filtering dropdown (e.g., docs.anthropic.com, github.com)
• Search functionality across chunk content
• Click-through navigation between chunks
• Expandable metadata sections

Integration Points

• Clickable document count badges in knowledge cards
• Browse button in knowledge item cards
• Modal overlay with smooth animations

Technical Requirements

• API: GET /api/knowledge-items/{source_id}/chunks with domain filtering
• Frontend: Modal component with filtering and search
• UI: Responsive design with proper scrolling
• Data: Preserve full chunk content and metadata

🔧 Implementation Notes

• Focus exclusively on browsing existing crawled content
• Clean separation from upload functionality (separate feature)
• Reuse existing UI patterns (similar to CodeViewer)
• Server-side domain filtering for performance

✅ Acceptance Criteria

• Modal opens when clicking document count badge
• Domain filtering works correctly
• Search filters content in real-time
• Chunk navigation is smooth and intuitive
• Metadata displays in expandable sections
• Responsive design works on different screen sizes
• Clean integration with existing knowledge base UI

🔗 Related

This feature complements the existing knowledge base by providing detailed inspection capabilities for crawled content.

This issue tracks the document browsing functionality only. Document upload features are handled separately.