LLMsTxt Generator

# llms.txt Generator Extension scans your `sitemap.xml`, convert pages & live sites to LLM-optimized Markdown, and export instantly. It generates a single llms.txt file and multiple llms-full.txt file for each page which includes all the links. It automatically downloads the whole generated markdowns as a single .zip file. --- ## Key Features - **Recursive Sitemap Scanning** - Parses your `sitemap.xml` and any nested sitemaps, following only valid `http(s)` URLs. - Filters out non-HTTP links for focused scanning. - Generates a single llms.txt file along with sub pages for more details - Zips all generated files and auto downloads as a single file once scan is completed - **Markdown Export (LLMsTxt Format)** - Converts HTML pages into clean **ATX-style headings** (`#`, `##`, …), fenced code blocks, and absolute URLs. - Removes `<script>`, `<style>`, and `<button>` tags; - Resolves relative links and images to **full URLs** for seamless static `llms.txt` content generation. - **Current Page Converter** - One-click “Convert Current Page” grabs the **rendered DOM** (supports SPA/React/Vue content). - Prepends `<title>` as `# Heading` and `<meta name="description">` as `> Blockquote`. - Ideal for ad-hoc page audits, AI training data extraction, and quick Markdown previews. - **Embed & SEO Metadata Guidance** - Built-in **Embed** tab with snippets: ```html <link rel="alternate" type="text/llmtxt" href="https://example.com/llms-full.txt" title="LLMsTxt version" /> <meta name="llmtxt" content="https://example.com/llms-full.txt" /> ``` - Publish `llms-full.txt` files alongside your pages for easy LLM ingestion and SEO signals. - **Intuitive Modern UI** - Four tabs: **Generator**, **Current Page**, **Embed**, **About**. - Real-time **progress bar** & **auto-scrolling log**. - ⚠️ User warning prevents accidental closure during scanning. - **Copy to Clipboard** for instant Markdown transfer. - **Privacy-First & Offline-Capable** - 100% local conversion—no external servers, no tracking. - Uses Chrome MV3 Offscreen API for accurate DOM parsing. --- ## How It Works 1. **Auto-Detect** your sitemap URL (`https://your-site.com/sitemap.xml`) on secure pages. 2. **Offscreen Rendering** fetches pages in a hidden DOM, executing scripts for dynamic content. 3. **Clean & Normalize** HTML: strip unwanted nodes, normalize whitespace per text node. 4. **Convert to Markdown** with Turndown: - Headings → `#`–`######` - Code → `lang …` - Links → `[text](absolute-url)` - Images → `` - JSON-LD → `application/ld+json …` 5. **Download or Copy** your domain’s ZIP or current-page Markdown. Please review the downloaded content before publishing under your webpage. If you see something unwanted, please report it as issue on open-source repository https://github.com/plainsignal/llmstxt --- ## Why Choose LLMsTxt Generator? - **SEO & Content Marketing**: Ideal for content audits, static migrations, UTM tracking, and structured data extraction. - **AI, LLM Workflows & RAG**: Prep training data, generate knowledge bases, accelerate AI-driven insights. - **Developer Productivity**: Integrates with CI pipelines, GitHub Actions, and static site generators. - **Flexibility & Extensibility**: Open-source under Apache 2.0 — https://github.com/plainsignal/llmstxt and contribute! It is community driven extension. Open an issue if something is not working as expected.