A bundle of utilities to extract documentation content from DeepWiki for LLM consumption.
- Copy content formatted for LLMs
- Download content as markdown
- Automatically ignores script tags and non-content elements
- Focuses on main content area
- Extracts navigation structure
- CLI tool with Playwright for automated full-site extraction
- Install in TamperMonkey/GreaseMonkey
- Automatically adds extraction buttons to documentation pages
- Location:
tampermonkey/deepwiki-extractor.user.js
- Automated extraction of entire documentation sites
- Uses Playwright to avoid CORS issues
- Generates a single markdown file with all pages
- Location:
cli/ - Best for: Extracting complete documentation sites automatically
cd cli
npm install
npm run install-browsersUsage:
# Extract entire documentation site
npm run extract -- https://deepwiki.com/nextjs/saas-starter/1-overview
# Extract single page
npm run extract -- https://example.com/docs/page --single -o output.mdSee cli/README.md for full documentation.
- Install TamperMonkey extension in your browser
- Open
tampermonkey/deepwiki-extractor.user.js - Click "Install"
All utilities add a floating button to the page that allows you to:
- Copy content to clipboard (formatted for LLMs)
- Download as markdown file
- View extracted navigation structure
- Add site-tool to extract documentation from DeepWiki
- Fix mermaid diagrams
- Add pdf export
MIT