Skip to main content

How It Works

Content Extraction​

The plugin intelligently extracts page content through several steps:

  1. Smart Content Selection — Automatically identifies the main documentation content area (main article or main .markdown)
  2. Clean Extraction — Removes navigation, sidebars, headers, footers, buttons, and other UI elements
  3. Markdown Conversion — Converts HTML to properly formatted markdown with preserved structure
  4. AI-Optimized Format — Includes page title, URL, and clean content

What Gets Extracted​

The conversion handles all common documentation elements:

  • Headings (h1-h6) with proper # syntax
  • Code blocks with language detection and syntax preservation
  • Inline code with backtick formatting
  • Tables with proper markdown table syntax
  • Lists (ordered and unordered)
  • Links with [text](url) format
  • Images with ![alt](src) format
  • Blockquotes and admonitions
  • Bold and italic text

Example Output​

When you click "Copy page" on a documentation page, you get clean markdown like this:

# Page Title

URL: https://your-site.com/docs/page

Your documentation content here, with all formatting
preserved as clean markdown...

## Section Heading

- List items preserved
- Links converted to [markdown](https://example.com)

Code blocks with language detection:

Injection Strategy​

The plugin uses a multi-tier injection system:

  1. Fast injection — For SPA navigation when the sidebar is already in the DOM
  2. Reliable injection — For page refreshes with retry logic (up to 3 seconds)
  3. Fallback injection — For pages without a table of contents, positions in the article area

The button automatically adapts to route changes, window resizes, and mobile/desktop layout transitions.