Built your site with Lovable, Bolt or another AI tool? These vibe coding SEO checks show if Google can read and index it, and how to fix what’s missing.
- Ammar Saleem
- Posted on
Table of Contents
Google says websites don’t need llms.txt, content chunking or AI-specific schema to appear in AI Overviews or AI Mode. None of these tactics gives a page a special ranking advantage in Google Search. An llms.txt file can still be useful for developer tools and AI agents that support the format, but it isn’t a shortcut to AI visibility.
Google has been unusually direct about several AI SEO tactics doing the rounds. Some have become familiar line items in agency proposals and technical audits, yet Google’s guidance gives website owners little reason to prioritise them.
Take llms.txt SEO. It has attracted plenty of attention as a way to make websites easier for AI systems to understand. But does Google’s AI search actually use it? And if Google says a tactic offers no special advantage, why does it keep showing up on SEO roadmaps?
This guide separates Google’s documented recommendations from industry assumptions. We’ll cover what the llms.txt file is, whether llms.txt works, where it may still be useful, and how much attention content chunking and AI schema deserve.
What Is an llms.txt File?
An llms.txt file is a Markdown-formatted text file placed in your website’s root directory. It gives compatible AI tools a structured guide to your website, listing important pages alongside short descriptions. Jeremy Howard of Answer.AI proposed the llms.txt format in 2024.
The format is deliberately simple. It starts with an H1 containing your website’s name, followed by a short summary. H2 headings group links into sections, and each link has a brief description. An optional companion file, llms-full.txt, can hold the full text of your key pages.
Unlike robots.txt, which controls crawler access, llms.txt only points compatible tools towards selected information. It has no say over which pages crawlers can or can’t access.
What Is llms.txt in SEO?
In SEO, llms.txt is usually pitched as a generative engine optimisation (GEO) tactic. The theory is that a curated guide to your website makes your content easier for AI tools to find and process.
The evidence doesn’t back that up for search. Google has confirmed its search systems, including AI Overviews and AI Mode, do not use llms.txt. OpenAI, Anthropic and Perplexity haven’t publicly documented using third-party llms.txt files in their crawler or search documentation either. That leaves little support for claims that publishing one improves AI search visibility.
Adoption figures add another angle. A June 2026 analysis of the top 10,000 websites found that 5.61% had a valid llms.txt file. Shopify also appears to have lifted adoption after adding the file to its stores by default in 2026. Still, adoption tells you how many sites have the file, not whether it does anything.
For SEO teams, the practical question is what the file actually achieves. Its real value lies with compatible tools and agent workflows that use it as a shortcut to relevant website information.
Does llms.txt Work?
It depends on what’s reading the file. For Google Search, llms.txt offers no advantage. Google doesn’t use it for AI Overviews, AI Mode or any other Search feature, so publishing one will neither improve nor harm your Google rankings.
Other AI platforms are less clear-cut, but none of the major players has made llms.txt a standard requirement for their AI search or crawler systems.
| System | Does it use llms.txt? | What to expect |
|---|---|---|
| Google Search, AI Overviews and AI Mode | No, per Google’s guidance | No ranking or citation effect |
| ChatGPT, Claude, Gemini and Perplexity | No established general support | No measurable citation lift in studies |
| Coding agents such as Cursor and Windsurf | Can use it when supplied as a documentation source | Useful for technical documentation and API context |
| Chrome Lighthouse agentic audits | Checked as optional | Optional machine-readability check |
The verdict is simple. llms.txt gives you no ranking advantage in Google Search, and research has yet to show a meaningful lift in AI search citations. It can still provide useful context when a developer tool or agent is set up to read it.
For most websites, that makes it an optional technical extra rather than a priority GEO investment.
Do Content Chunking and AI Schema Help You Rank in AI Search?
llms.txt SEO is one of three AI SEO tactics Google addresses in its guidance on AI features in Search. Content chunking and AI-specific schema come with similar promises of better AI visibility. Google’s position on both is clear: neither needs a special optimisation strategy.
Does Content Chunking Work?
Content chunking means breaking a page into small, self-contained sections so AI systems can retrieve and reuse individual passages. Google says you don’t need to restructure your content this way for AI search.
Its systems can understand context and nuance across a whole page, then pick out the information relevant to a search. Google’s AI features also use query fan-out, running multiple related searches to explore different parts of a question. That gives them more ways to find relevant information across the web.
Descriptive headings and well-organised paragraphs still make content easier to read and understand. Structure your pages around the reader’s questions and the complexity of the topic. There’s no ideal paragraph length, though, and no need to turn every section into an isolated answer.
Does AI Schema Work?
Google doesn’t require structured data for inclusion in AI Overviews or AI Mode, and there’s no special schema.org markup that unlocks these features.
That doesn’t make schema irrelevant. Structured data helps search engines interpret your pages and can make them eligible for rich results. Keep using relevant schema that accurately reflects your visible content, but don’t treat it as a shortcut to AI citations.
What earns citations is content competitors can’t copy. Original research, first-hand data and expert opinion create non-commodity content that AI answers have a reason to cite.
How Do You Create an llms.txt File?
Choose your most useful pages, add a short description for each, and publish the file at yourdomain.com/llms.txt. For most websites, setup takes less than an hour.
- Audit your site and choose 10 to 30 priority pages.
- Write a plain, one-line description for each page.
- Group the links under H2 headings such as Services, Guides and About.
- Create the file by hand or with a plugin. Yoast SEO, All in One SEO and Rank Math can all generate one.
- Upload it to your root directory and confirm it loads in a browser.
Keep descriptions specific and accurate, and prioritise pages that are clearly relevant to your business. Review the file whenever you publish important content or change your site structure.
How Can You Check if Your llms.txt File Is Valid?
Visit yourdomain.com/llms.txt in your browser. The file should load as plain text and return an HTTP 200 status code. Then check that:
- The file sits in your root directory, not a subfolder.
- The structure follows Markdown format, with an H1, a summary and H2 sections.
- Every link resolves without redirects or errors.
- None of the listed pages is blocked from crawling or set to noindex.
- The listed URLs are current and consistent with your XML sitemap.
A valid file proves nothing about visibility. The more useful test is whether AI answers mention your brand at all, and a tool such as Brand Radar can track that directly. For Google’s own AI surfaces, the Google Search Console generative AI report is another useful place to look.
AEO vs SEO vs GEO: What Is SEO for LLMs?
It’s still SEO, with a few extra considerations for how AI-powered search systems retrieve, interpret and present information. Optimising for generative AI search builds on established search practice rather than replacing it with a new discipline.
Here’s how AEO vs SEO generative engine optimisation breaks down:
- SEO earns rankings for your pages in search results.
- AEO (answer engine optimisation) shapes content so it can serve as the direct answer to a question.
- GEO (generative engine optimisation) aims to get your brand cited inside AI-generated responses.
The overlap is large. Strong rankings feed AI citations, and clear answers help both. Queries also behave differently as AI changes how search works, because AI Overviews change how each keyword type performs. Informational searches feel the shift first.
The tactics that hold up against every AI SEO myth are familiar ones:
- Lead each section with a direct answer.
- Publish original insight backed by real examples.
- Keep your site fast, crawlable and logically linked.
- Earn relevant mentions through useful content and genuine industry contributions.
Key Takeaways
- Google does not require llms.txt for inclusion in AI Overviews or AI Mode.
- Content chunking and AI-only schema carry no special ranking benefit in Google Search.
- llms.txt may still help compatible developer tools and AI agents access website information.
- AEO and GEO still depend heavily on established SEO fundamentals.
Final Remarks
llms.txt, content chunking and AI schema share one flaw: each promises a shortcut that Google says doesn’t exist. llms.txt SEO is harmless if a plugin builds the file in minutes, but it’s a poor place to spend strategy hours.
The better path is a site that answers questions well and earns trust. Adcept Marketing can audit your site, separate the AI tactics worth testing from the ones to drop, and build content that both search engines and AI answers reward.
FAQs
Does an llms.txt File Block AI Crawlers?
No. Blocking is the job of robots.txt, which lets you allow or disallow bots such as GPTBot or ClaudeBot. llms.txt only suggests pages and has no enforcement power over any crawler.
Is llms-full.txt Different From llms.txt?
Yes. llms-full.txt bundles the full text of your key pages into one Markdown file. It suits documentation-heavy sites where agents need complete context. Google Search doesn’t use it either.
Can llms.txt Hurt My SEO?
There’s no evidence it does. Google ignores the file for ranking purposes. The only real risk is listing outdated or sensitive pages, so review the file whenever your site changes.
Should a Small Business Website Bother With llms.txt?
Only once the fundamentals are in place. A plugin can create the file in minutes, which makes it a low-cost extra. Clear service pages, local signals and original content will move results far more.
Ammar Saleem is a Copywriter at Adcept Marketing who’s spent the last five years helping brands turn smart automation into real results. From search engine optimization to sales funnels and landing pages, he creates content that connects with audiences and drives action. Ammar Saleem has a talent for breaking down complex ideas into clear, practical messaging and he loves helping businesses simplify their marketing so growth feels effortless.
AI-powered marketing. Eight channels. One customer journey.
Related Insights
Built your site with Lovable, Bolt or another AI tool? These vibe coding SEO checks show if Google can read […]
- AI Work, Digital Marketing
Scaled content abuse is Google’s rule against mass-made pages built for rankings, not people. See when AI content crosses the line and how to stay safe.
Scaled content abuse is Google’s rule against mass-made pages built for rankings, not people. See when AI content crosses the […]
- AI Work, Digital Marketing
Can you block AI crawlers? Yes, with robots.txt, but it’s a request, not a lock. See which bots to block, which to keep and the new AI training controls.
Can you block AI crawlers? Yes, with robots.txt, but it’s a request, not a lock. See which bots to block, […]
- AI Work, Digital Marketing
Learn how to track AI traffic in GA4, why ChatGPT visits can show as Direct, and how to see Google AI Mode clicks. Simple steps for Australian businesses.
Learn how to track AI traffic in GA4, why ChatGPT visits can show as Direct, and how to see Google […]
- AI Work, Digital Marketing