llms.txt & llms-full.txt Generator, Validator & Sitemap Importer

When AI models (such as ChatGPT Search, Perplexity AI, Claude Search, and NotebookLM) search the web to answer user questions about your product, library, or company, they face heavy HTML boilerplate (scripts, navbars, ads, trackers) that dilutes context and wastes tokens.

`llms.txt` (specified at llmstxt.org) is the standardized markdown file hosted at `https://yourdomain.com/llms.txt` (similar to `robots.txt`) that provides AI search crawlers with a clean, curated markdown roadmap of your most important documentation and guides.

A valid `llms.txt` file consists of structured markdown sections: 1. Top-level H1 title (`# Project Name`) 2. Summary Blockquote (`> Concise 1-2 sentence description of what the site is`) 3. Curated Sections (`## Core Documentation`, `## API Reference`) 4. Link list with markdown format: `- [Resource Title](https://domain.com/doc.md): Description of why this page matters.` 5. Optional "Optional" subsection for secondary resources (`### Optional`).

- **`/llms.txt`:** A lightweight, token-efficient index (typically 500 to 2,000 tokens) pointing to clean markdown URLs. - **`/llms-full.txt`:** A single, concatenated bundle containing the entire raw text of all documentation pages, ideal for feeding an entire library into an LLM context window in a single request.

Our studio features a browser-native `sitemap.xml` parser. Paste or upload your XML sitemap, select the key documentation URLs, and generate a validated `llms.txt` file in seconds. The live token counter ensures your index stays lean and fits within the lowest LLM token budgets.