Skip to main content
Zark processes documents and images to understand their content and make it searchable.

How Content Processing Works

When content is processed, Zark:
  • Extracts text and structure from documents
  • Analyzes images and visual content
  • Identifies key entities, topics, and sections
  • Indexes content for fast retrieval
  • Prepares data for AI-powered search and analysis
This process allows Zark to understand files beyond filenames or folders.

What Gets Indexed

Content Library indexes:
  • Text from documents and presentations
  • Headings and section structure
  • Tables and structured data
  • Key terms, entities, and concepts
  • Text detected inside images
  • Visual attributes of images
This indexed data powers search, filtering, and AI analysis.

Supported Content Types

Content TypeFormats
DocumentsPDF, Word, Markdown, HTML, presentations, and plain text files
ImagesPNG, JPG, WebP, SVG, screenshots, and diagrams
Code FilesCommon programming and configuration formats
Each file type is processed to preserve structure and meaning.

How AI Interprets Content

AI interpretation enables:
  • Understanding meaning, not just keywords
  • Finding content by concept
  • Comparing content across files
  • Answering questions about content
See How Indexing Works for how content is indexed.

What Becomes Searchable

After processing, content becomes searchable by:
  • Topic or concept
  • Entities mentioned
  • Visual descriptions
  • Specific sections or data points
Search works across all processed content. See Search & Discovery for how search works.