llms.txt Generator and AI Access
Can AI bots see your site? The draft is ready, the permissions are visible.
Draft and summary:
What is llms.txt?
This tool does two jobs at once: it reads the permission state for GPTBot, ClaudeBot, PerplexityBot, Google-Extended and OAI-SearchBot/Bingbot in your robots.txt, and it builds an llms.txt draft from your sitemap. The draft is previewed and downloaded; it is uploaded to the root as /llms.txt.
llms.txt is a plain-text file at the root of a site that came out of the llmstxt.org proposal. In Markdown it tells AI models what the site is and where the key pages are: a title, a short description and link lists in sections. Google does not treat llms.txt as an official ranking signal; still, for answer engines and agents it is a readable map of the site.
The file's value is in its content: every link needs a one-sentence description of the page, the sections should mirror the site's real structure and the list should stay current. That is why the tool builds the draft from the real titles and descriptions of the pages in your sitemap.
Example llms.txt
# Example Co
> Ankara-based SEO and GEO agency; technical SEO, content, backlinks.
## Services
- [SEO Packages](https://example.com/seo-packages): Monthly SEO packages and scope.
- [Backlink Packages](https://example.com/backlinks): Manual backlink packages.
## Blog
- [What is SEO?](https://example.com/blog/what-is-seo): A starter guide.Where do AI bots stand in robots.txt?
Blocking AI bots means not appearing in AI answers. robots.txt can carry separate rules for GPTBot (OpenAI), ClaudeBot (Anthropic), PerplexityBot, Google-Extended (Gemini) and OAI-SearchBot / Bingbot (ChatGPT search and Copilot). The tool shows one of three states per bot: allowed, partial (some paths closed) or blocked; without a bot-specific rule, the general (*) rule applies.
Some sites close training crawlers (GPTBot, Google-Extended) while keeping search bots (OAI-SearchBot) open. That can be a deliberate choice; what matters is knowing the decision and not closing content directories by accident.
Five bots
- GPTBot: OpenAI's crawler (training and browsing)
- ClaudeBot: Anthropic's crawler
- PerplexityBot: the Perplexity answer engine
- Google-Extended: content use permission for Gemini
- OAI-SearchBot / Bingbot: ChatGPT search and Microsoft Copilot
User-agent: GPTBot
Allow: /
How the draft is prepared and where it goes
The draft reads up to 25 pages from the Sitemap line in robots.txt (or /sitemap.xml if there is none), takes every page's title and meta description and groups them by the first path segment. Without a sitemap the homepage links are used. The generated file is downloaded and uploaded to the root as llms.txt; section names and descriptions can be edited by hand if needed.
- 1
robots.txt and the sitemap are read
Without a Sitemap line, /sitemap.xml is tried; up to 25 pages.
- 2
Titles and descriptions are collected
Every page’s title and meta description; sections by the first path segment.
- 3
The file is downloaded and uploaded to the root
It should open at https://yourdomain.com/llms.txt; the text can be edited by hand.
Does llms.txt affect SEO?
llms.txt is not required for SEO and does not directly change Google rankings. For GEO it is a low-cost step: AI agents can draw on the file when they summarise the site. The real effect comes together with robots.txt permissions, structured data and answerable content; the SEO + GEO Readiness Score tool measures those three.
Another check to run? Our free tools are ready.
- SEO CheckerSEO + GEO Readiness Score, 303 rules
- Google Rank CheckerRank and address, top 100
- Keyword AnalysisVolume, difficulty, ideas and package
- AI Visibility CheckerYour brand in ChatGPT, Gemini and Claude answers
- Schema CheckerJSON-LD gaps, per type
- DA PA CheckerDA, PA and Spam Score
- KDV CalculatorKDV inclusive, exclusive, base and withholding
- Discount CalculatorPercent off, rate, reverse and chained discounts
- Percentage CalculatorPercent of, what percent, change and raise
- WhatsApp Link Generatorwa.me link with a message, and a QR code
Being cited in AI answers as a source.
Bot permissions are the first step; being a source takes schema, answerable content and trust signals. GEO is part of every SEO package.
- GEO in every package
- Monthly report
- 11+ years
llms.txt: no question left unanswered.
From llms.txt to GPTBot: a short answer to every question.
Generate llms.txtllms.txt is a plain-text file at the root of a site that summarises its content for AI models in Markdown. It holds a title, a short description and link lists split into sections; it helps answer engines and agents understand the site.
Google has said it does not treat llms.txt as an official ranking signal. The file does not change Google rankings; it offers a readable site map for AI agents and answer engines.
Add a User-agent: GPTBot line to robots.txt followed by Disallow: / (block) or Allow: / (allow). Without a bot-specific rule, the general User-agent: * rules apply. The same structure works for ClaudeBot, PerplexityBot and Google-Extended.
This tool shows the permission state for GPTBot and OAI-SearchBot in robots.txt. To see actual crawling, look for the GPTBot, ChatGPT-User and OAI-SearchBot user agents in your server logs.
The draft reads up to 25 pages from the sitemap. On large sites we suggest hand-picking the most important pages and sections and extending the file; it is plain text and can be edited in any editor.
At the root, next to robots.txt: it should open at https://yourdomain.com/llms.txt. Optionally an llms-full.txt file can be added for the full-text version.
As long as Googlebot and Bingbot stay open, classic rankings are unaffected. When GPTBot, ClaudeBot and PerplexityBot are blocked, the chance of being cited in those assistants' answers drops; in other words, GEO visibility shrinks.
Need help?
Reach out, we are here 24/7.
For your questions, project or ideas, our expert team is with you 24/7. The relevant team reviews your request and gets back to you.