Documentation and developer sites
, where assistants answer technical questions.
Use this llms.txt generator to build a plain-text Markdown file that lists your site name, a short summary and your most important pages, each with a one-line description. Fill in the form, click Generate, copy the result and save it as llms.txt at your domain root. It takes about two minutes. Read the section on limits below before you expect results: llms.txt is a proposal, not an official ranking factor.
Generate your llms.txt fileAdd your most important pages. AI models will prioritise these when citing your site.
llms.txtllms.txt and upload it to your server root so it is accessible at yourdomain.com/llms.txt. On WordPress, upload via FTP or File Manager to the public_html folder.
llms.txt is a Markdown file placed at the root of a website, for example yoursite.com/llms.txt, that gives large language models a short, curated guide to the site: a title, a summary and links to the pages that matter most. Jeremy Howard of Answer.AI proposed it in September 2024. It is a voluntary convention, not an official standard, and no search engine has confirmed it as a ranking factor.
Language models read the web in small windows. A model that visits a typical site meets menus, scripts, ads and long pages that are hard to turn into clean text. The llms.txt proposal offers a shortcut: a single file that says what the site is and where the best, cleanest pages are, written in Markdown that both people and machines can read.
The idea is most useful for documentation, developer sites and product knowledge bases, where an AI assistant helping a user needs the right page quickly. For a blog, agency or shop it is a low-cost tidy-up with unproven returns. That is the honest position, and we say it plainly because several pages on this topic oversell it.
Not in any way Google has confirmed. Google has not said that it reads llms.txt for Search or AI Overviews, and the file is not a ranking factor. Google's own guidance says no special markup is needed to appear in AI features. Other AI providers have not published formal commitments to follow the file either. Some AI tools and developer agents do read it, mostly for documentation sites, so it can help when a person points an assistant at your domain.
So why create one? Because it costs little, it forces you to decide which ten or twenty pages best describe your business, and that exercise improves your internal linking and your own content planning. Treat any traffic gain as a bonus, not a promise.
The proposed format is simple Markdown. The table shows each part and what to write.
| Part | Markdown element | What to put there |
|---|---|---|
| Site title | H1 heading (required) | Your site or brand name |
| Summary | Blockquote under the title | A short, factual description of the site and who it is for |
| Notes | Plain paragraphs (optional) | Extra context a model should know, such as scope or language |
| Sections | H2 headings | Groups such as Guides, Tools, Products or Company |
| Page links | List items with link and a note | Page title, URL and one line on what the page covers |
| Secondary pages | An H2 named Optional | Pages a model may skip when it needs a shorter context |
Write descriptions as facts, not slogans. "Step-by-step guide to setting up GST invoicing in Tally" helps a reader or a model. "The best guide ever" does not.
| File | Job | What it controls | Status |
|---|---|---|---|
| robots.txt | Tells crawlers what they may fetch | Crawler access by path | Established, RFC 9309 |
| sitemap.xml | Lists URLs for discovery | Which URLs exist and when they changed | Established, supported by search engines |
| llms.txt | Guides models to key pages | Nothing enforceable, only a suggestion | Community proposal, 2024 |
The three files do different jobs and none replaces another. Blocking an AI crawler requires robots.txt rules, or server-level controls for crawlers that ignore them. Listing a page in llms.txt does not unblock it. Build the access rules with the Robots.txt Generator and the discovery list with the XML Sitemap Generator.
There is no independently verified list. Some AI developer tools and documentation platforms read the file. Whether a given crawler such as GPTBot, ClaudeBot or PerplexityBot parses it and changes its behaviour is not confirmed in public documentation we can cite. Check each provider's current crawler documentation before you rely on it, and watch your server logs to see who actually requests the file.
, where assistants answer technical questions.
That want assistants to point to current pricing, docs and support pages.
Who want to name their flagship guides.
Who want a short, accurate service summary in one place.
With a stable set of category and policy pages.
Update the file when you publish a flagship page, restructure the site or retire an important URL. A review every quarter is a reasonable habit for most sites. Rerun the generator, compare the output with your live file and replace it.
An llms.txt file is public by design, so never include private URLs, staging links or internal notes. The tool processes your input in the browser. Use the generator for public information only.
It is a Markdown text file at your domain root that gives language models a short summary of your site and links to its most important pages. Jeremy Howard of Answer.AI proposed it in September 2024.
No. It is a community proposal, not a W3C or IETF standard, and no search engine has confirmed it as a ranking factor. Treat it as an optional, low-cost aid.
Google has not confirmed that it reads llms.txt for Search or AI Overviews. Google says no special markup is required to appear in its AI features, so focus on content quality and clear structure.
Upload it to the root of your domain so it loads at https://yourdomain.com/llms.txt. It must be publicly reachable with no login and no redirect. On WordPress, use your host's file manager or FTP.
robots.txt tells crawlers which URLs they may fetch. llms.txt only suggests which pages are most useful and enforces nothing. Use robots.txt to control access and llms.txt as an optional guide.
Include your site name, a short factual summary and links to ten to twenty key pages with one-line descriptions. Group them under clear headings and put lower-priority pages under an Optional section.
No. It cannot block anything. To block AI training or search crawlers, add rules to robots.txt, and use server or firewall rules for crawlers that ignore robots.txt.
A sitemap lists every URL for discovery. llms.txt is a short, curated, human-readable summary. They do different jobs, and having a sitemap does not make llms.txt redundant or required.
Update it when you publish a flagship page, restructure your site or remove a listed page. A quarterly review works for most sites.
Yes. It is free, with no signup. You enter your details, copy the output and upload the file yourself.
Preview your title and description in Google on desktop and mobile.
Open the toolCreate JSON-LD for 16 page types and check which still earn rich results.
Open the toolCheck 20 on-page trust signals and get a prioritised fix list.
Open the toolNext step: Open the llms.txt Generator, list your ten best pages and publish the file. Then check your answer-first structure with the AEO Readiness Checker.
Open the llms.txt GeneratorOriginal, well researched guides are always welcome here.
A guest posting platform where analysts, founders and specialists publish original, well researched guides.
connect@writorybuzz.comNo ratings yet. Be the first.