Free · No sign-up · Runs in your browser
llms.txt generator and validator
Create an llms.txt file from a simple form, or paste yours to check it against the llmstxt.org format. It's a cheap way to give AI agents a map of your site, not a ranking lever.
Your llms.txt
# Brightdesk > Brightdesk is help desk software for support teams of 2 to 50 people. Plans start at $12 per agent per month. Brightdesk includes a shared inbox, live chat, a knowledge base and SLA reports. It integrates with Gmail, Slack and Shopify. ## Docs - [Getting started](https://brightdesk.example/docs/getting-started): set up your inbox and invite agents - [API reference](https://brightdesk.example/docs/api): REST endpoints, authentication, rate limits - [Integrations](https://brightdesk.example/docs/integrations): Gmail, Slack and Shopify setup ## Product - [Features](https://brightdesk.example/features): shared inbox, live chat, knowledge base - [Pricing](https://brightdesk.example/pricing): plans, agent limits, annual discount ## Guides - [How to set SLAs for a small support team](https://brightdesk.example/blog/sla-small-team): response targets and examples - [Brightdesk vs Zendesk](https://brightdesk.example/compare/zendesk): features and pricing side by side ## Optional - [Changelog](https://brightdesk.example/changelog) - [About us](https://brightdesk.example/about)
Passes all checks · Sections: 4 · Links: 9
Checked in your browser against the llmstxt.org format. Nothing is sent anywhere.
Checklist
Valid, with suggestionsPassed 4Warnings 6Errors 0
Pass: H1 title: Brightdesk
The file starts with a single H1, the only required element of the format.
Warning: No blockquote summary
Recommended: right after the H1, add a line starting with > with one sentence on what the site is and who it's for.
Pass: Length: 319 characters
Above the 50-character minimum. The Lighthouse audit rejects files that are too short.
Pass: Markdown links: 3
The Lighthouse audit expects at least one link written [title](url).
Pass: H2 sections: 2
Links are grouped under H2 headings, as the format expects.
Warning: Lines off format: 1
Inside H2 sections the format expects list items written - [Title](url), with an optional note after a colon. Check lines 11.
Warning: Empty sections: 1
These H2 sections have no links: "Blog". Add links or remove them.
Warning: Relative or non-web URLs: 1
Prefer full URLs that an agent can follow from anywhere: "docs/api".
Warning: Duplicate links: 1
Listed more than once: "https://brightdesk.example/docs/getting-started". Keep each URL once, in its best section.
Warning: H3 or deeper headings: 1
The format only uses an H1 and H2 sections. Lines 10 use deeper levels: make them H2 sections or plain text.
Pass:
Warning:
Error:
llms.txt describes your site to machines. Whether assistants actually recommend you is something you measure.
Track this in ChatGPT, Gemini, Perplexity… with Citegram (free)How to create an llms.txt file with this generator
1. Name your site. The name becomes the H1, the only element the llmstxt.org proposal requires.
2. Write a one-sentence summary. It becomes the blockquote under the title. Say what you sell, to whom and, if you can, at what price. Plain facts work better than a tagline.
3. Add details if needed. A short paragraph with your category, main features or integrations. Leave it empty if the summary says enough.
4. Build your sections. Each section becomes an H2 with a list of links written - [Title](URL): note. Typical sections are Docs, Product and Guides. Use full URLs (https://…) so an agent can follow them from anywhere.
5. Turn on the Optional section for secondary pages such as a changelog or an about page. The proposal reserves ## Optional for links an agent can skip when it needs a shorter context.
6. Copy or download the file and publish it at the root of your site, at /llms.txt. The line under the output tells you whether the result passes the validator's checks.
How to validate an existing llms.txt
Open the Validate tab and paste the content of your file. The checklist updates as you type. Nothing is sent anywhere: the check runs in your browser.
Each line is a pass, a warning or a fail. A fail means the file breaks a rule of the format or one of the checks in Chrome's Lighthouse llms.txt audit: a missing or misplaced H1, no Markdown link, a file that's too short, or a URL that can't be followed. A warning means the file works but drifts from the recommended structure, for example a missing summary, an empty section, a duplicate link or an H3 heading.
The tool checks the text, not your server. After publishing, open https://yoursite.com/llms.txt to confirm it returns the file, then run the Lighthouse Agentic Browsing audit.
What each part of the file means
The proposal defines a fixed order: an H1 with the site or project name, a blockquote summary, optional free text without headings, then H2 sections that list links. The goal is a compact starting point an AI agent can read instead of parsing heavy HTML pages.
The H1 names the site. The blockquote gives the context needed to understand the rest. The H2 sections point to the pages that matter, each with a short note on what the page contains. The Optional section holds links that can be dropped first.
The proposal also suggests offering clean Markdown versions of important pages at the same URL with .md appended. That's optional and mostly useful for documentation sites.
What llms.txt does and doesn't do
llms.txt is a community proposal published in September 2024, not an official standard. Google states that Google Search, including AI Overviews and AI Mode, ignores it, and that adding one neither harms nor helps your rankings. OpenAI, Anthropic and Perplexity don't document their assistants reading it to choose which sources to cite.
It also controls nothing: it contains no directives and can't block crawling or AI training. That is the job of robots.txt, covered in our guide to AI crawlers and robots.txt.
Where it helps is with agents and developer tools. Documentation sites adopted it first, because developers point coding assistants at docs and a clean index saves those tools from crawling the whole site. For a marketing site, see it as a cheap, low-stakes addition. Our full guide to llms.txt covers the evidence in detail.
Best practices for a useful llms.txt
Keep it short and current. A dozen well-chosen links beat a copy of your sitemap. A file with old pricing or dead links is worse than none, so skip it if nobody will maintain it.
Write notes that describe the page. "Plans, agent limits, annual discount" tells an agent more than "Learn more".
Generate it at build time if you can, from the same list of pages your site already uses, so it never drifts. Citegram's own site publishes an llms.txt generated that way.
Put your effort where answers come from. Clear facts on the pages assistants read and coverage on third-party sites matter more. We explain why in how AI assistants choose the sources they cite.
Measure whether assistants recommend you
This tool can tell you your file is well formed. It can't tell you whether ChatGPT or Perplexity name your brand when a buyer asks for a recommendation. That you have to measure.
Citegram is a free Chrome extension that asks your buyers' questions in ChatGPT, Gemini, Perplexity, Claude, DeepSeek, Grok and Google AI Overviews, using the accounts already signed in to your browser. For each answer it records whether your brand is mentioned, its position and the sources cited. Run a baseline before you publish llms.txt or any other change, then rerun the same prompts and compare trends. The free plan tracks one prompt; Pro adds unlimited prompts for $10 a month.
FAQ
How do I create an llms.txt file?
Fill in the generator above: a site name for the H1, a one-sentence summary, then H2 sections with links to your key pages. Download the result and upload it to the root of your site so it's reachable at /llms.txt. On a static site or framework, you can also generate it at build time.
Where do I put the llms.txt file?
At the root of your domain, like robots.txt: https://example.com/llms.txt. The proposal also allows a subpath for documentation, such as /docs/llms.txt. Check that the URL returns the file with a 200 status.
Does llms.txt help SEO or AI visibility?
Not in Google: Google says Search ignores llms.txt and that it neither helps nor hurts rankings. No major assistant documents using it to choose sources either. It can help AI agents and developer tools find your key pages, so treat it as a small addition, not a ranking lever. Read more in our llms.txt guide.
What's the difference between llms.txt and llms-full.txt?
llms.txt is a short index with links, as defined at llmstxt.org. llms-full.txt is not part of the proposal: it's a companion file some sites publish with their full content in one document, mostly for developer documentation.
How can I check that my llms.txt is valid?
Paste it in the Validate tab above to check it against the format: one H1 at the top, a blockquote summary, H2 sections of Markdown links, valid URLs. Then run Chrome Lighthouse's Agentic Browsing audit on your live site, which checks that the file is served and well formed.