A lightweight, high-performance, and privacy-first web tool designed to instantly generate standard /llms.txt files for websites. This tool is built entirely with pure HTML, CSS, and Vanilla JavaScript, ensuring 100% client-side execution with no server-side overhead and complete privacy.
Live Demo: https://heretool.com/llmstxt-generator
The llms.txt file is an emerging web standard proposed to help Large Language Models (LLMs), AI agents, and custom search crawlers understand a website's purpose, core features, and primary documentation paths efficiently. By serving a concise, structured Markdown-based text file at the root directory (/llms.txt), you can significantly optimize how AI models consume and contextualize your project.
-
🌐 Powerful Sitemap Import & Parsing
- Fetch URLs automatically via an optimized XML parser.
- Built-in fallback CORS proxies (
allorigins) for direct client-side fetching. - Manual copy-paste XML parsing support if server endpoints restrict direct requests.
- Batch select, filter, and quickly map URLs into beautiful structured list entries.
-
🏷️ Site Identity & Core Profiling
- Quickly configure Site Name, Canonical Root URL, and one-line summaries optimized for systemic token parsing.
- Dedicated Core Information block utilizing standard clean Markdown parsing for quick summaries, API outlines, and important directory indexing.
-
🗂️ Dynamic Component Extensibility
- Optional Custom Sections: Add structured subheadings (e.g., FAQ, Design Principles, API Specifications) dynamically.
- Links / Resource Mapping: Generate descriptive Markdown reference link components on the fly (text, URL, and sub-descriptions).
-
⚡ Live Interactive Previews
- Real-time character counts and text output rendered instantly inside a clean dark-mode syntax terminal block.
- Local clipboard export and fast standalone file downloading capabilities natively managed by browser blobs.
-
🛡️ 100% Client-Side & Serverless
- Zero external database connections or analytical trackers, ensuring maximum site load speeds and strict personal data handling compliance.
This project embraces standard, lightweight, web engineering fundamentals without modern bloated dependency ecosystems:
- Frontend Environment: Native HTML5 & Custom Fluid Responsive CSS Media Queries.
- Script Engine: Native ECMAScript / Vanilla JS (No jQuery, React, Vue, or Webpack required).
- Styling: CSS Flex/Grid layout fallbacks mapping beautifully across mobile and wide-panel screen viewports.
- Clone or download this repository.
- Put the
index.htmlfile into your project folder. - Open
index.htmldirectly inside any modern desktop or mobile browser.
This project is open-source software licensed under the MIT License. See the LICENSE file for more information.