hidekazu-konishi.com

About This Tool

llms.txt is a proposed convention for publishing a curated, Markdown-formatted guide to your website at /llms.txt, so that large language models (LLMs) and AI tools can find your most useful content without parsing complex HTML. This tool does two things, entirely in your browser:

• Generator: builds a structurally correct llms.txt from your site name, summary, and link sections - including bulk import of page lists in TSV, CSV, or Markdown list format - and concatenates page contents into an llms-full.txt file.
• Validator: checks an existing llms.txt against the structure described in the proposal at llmstxt.org, reporting line-numbered results in three levels (Error, Warning, Info) and labeling every finding as either derived from the spec wording or a clearly marked recommendation of this tool.

The design is informed by first-hand experience publishing and maintaining llms.txt and llms-full.txt files for this site's own content.

Validation Levels

How to Use

Generate llms.txt

1. Enter Site Information: the site name becomes the H1 and the summary becomes the blockquote
2. Add Sections: create H2 sections such as "Docs", "Examples", or "Optional", and add one link per line
3. Bulk Import: paste a page list ("URL<TAB>title<TAB>description", CSV, or a Markdown list) and import it into a section
4. Generate: click "Generate llms.txt" or press Ctrl+Enter (Cmd+Enter), then copy or download the result
5. Check: click "Validate This Output" to run the generated file through the validator

Build llms-full.txt

1. Add Pages: paste each page's full Markdown content with its title
2. Build: the pages are concatenated with separators, optionally starting with your site name and summary

Validate llms.txt

1. Paste or Drop: paste llms.txt content into the text area, or drag and drop a .txt / .md file
2. Validate: review line-numbered Errors, Warnings, and Info notes, each labeled "spec" or "recommendation"

Features:

• 🔧 llms.txt Generation: Build a structurally correct llms.txt from form inputs - H1, blockquote summary, free-form details, and H2 link sections
• 📥 Bulk Page Import: Paste page lists as TSV ("URL<TAB>title<TAB>description"), CSV, Markdown lists, or plain URLs and route them into any section
• 📚 llms-full.txt Builder: Concatenate full page Markdown into a single llms-full.txt with separators (clearly labeled as a community convention)
• ✅ Structure Validation: Line-numbered Error / Warning / Info results based on the proposal at llmstxt.org
• 🏷️ Spec vs. Recommendation: Every finding is labeled as derived from the spec wording or as this tool's recommendation - no invented "rules"
• 🔍 Duplicate URL Detection: Flags the same URL listed more than once across all sections
• 📂 File Drag and Drop: Drop a .txt or .md file to validate it instantly
• 📋 Examples Included: Documentation-site and blog presets for the generator, plus an intentionally broken sample for the validator
• 🔒 Privacy First: All processing happens locally in your browser - no data sent to servers, no URLs fetched
• ⌨️ Keyboard Shortcut: Ctrl+Enter (Cmd+Enter) to generate or validate instantly

FAQ

What is llms.txt?

llms.txt is a proposal (published at llmstxt.org) to add a Markdown file at /llms.txt on a website that gives LLMs a concise, curated overview of the site: an H1 title, a blockquote summary, optional free-form details, and H2 sections containing lists of links in the form - [name](url): description. The proposal also suggests serving clean Markdown versions of pages at the same URL with .md appended.

Do AI crawlers actually use llms.txt?

Adoption is varied and still evolving. llms.txt is a community proposal - not a ratified standard - and AI vendors do not uniformly document whether their crawlers or assistants read it. Some developer-documentation platforms generate llms.txt files, and some AI tools can consume them when pointed at them, but publishing an llms.txt file does not guarantee that any AI system will read or cite your content. Treat it as a low-cost, low-risk addition rather than a ranking lever.

How is llms.txt different from robots.txt and sitemap.xml?

robots.txt tells crawlers what they may access; sitemap.xml enumerates URLs for indexing; llms.txt is different from both - it is a human-readable, curated guide that tells an LLM which pages matter most and what they contain, in a format that fits within limited context windows. The three files complement rather than replace each other.

Is llms-full.txt part of the specification?

No. The llms.txt proposal defines only llms.txt itself (it mentions context-expansion tooling such as the FastHTML project's llms-ctx.txt files as one example implementation). llms-full.txt - a single file concatenating the full Markdown content of your pages - is a community convention popularized by documentation sites. This tool builds it because it is widely used, and labels it accordingly.

Where should I place llms.txt and how often should I update it?

The proposal specifies the root path /llms.txt (a subpath is optionally allowed). Update it whenever the set of pages you want AI tools to see changes - for example, when you publish new articles or documentation. Regenerating it as part of your normal publishing workflow keeps it from going stale.

Important Notes

• llms.txt is a proposed convention, not a ratified standard. Adoption by AI crawlers varies and is evolving - providing llms.txt does not guarantee that AI systems will read or cite your content, and this tool makes no SEO or AI-search ranking claims.
• Validation follows the structure described in the llms.txt proposal. The proposal explicitly declares only the H1 required; everything else in the documented structure is reported as Warning or Info rather than Error, and this tool's own suggestions are labeled "recommendation".
• URL checks are syntax-only. This tool never fetches URLs, so it cannot tell you whether a linked page actually exists.
• Sources: the llms.txt proposal at llmstxt.org and its repository at github.com/AnswerDotAI/llms-txt.

Related Tools

• Meta Tag OGP Generator Tool - generate Open Graph, Twitter Card, and basic meta tags with live previews
• Schema.org JSON-LD Generator Tool - generate structured data markup with Google Rich Results preview
• Markdown Live Editor Tool - edit the Markdown for your llms-full.txt pages with a live preview
• Markdown HTML Converter Tool - convert existing HTML content into the Markdown that llms.txt links should point to