10 mins | 18 Mar 2026
 Revised_dd4e633e.jpeg)
Picture this: a potential client asks ChatGPT or Perplexity to suggest a web development agency in Mumbai. The AI thinks for a moment, pulls from what it knows about your website, and responds. But here's what it doesn't tell you: it may have ignored your most important pages — your case studies, your pricing guide, your best articles — entirely.
Not because the content was bad. But because AI systems don't crawl websites the way Google does. They scan, extract, and move on — often missing the depth that actually sells you.
That's the problem llms.txt was designed to solve. And in 2026, with AI-mediated discovery becoming a meaningful part of how Indian buyers find vendors, it's a problem worth understanding clearly.
This article gives you the honest truth about llms.txt — what it is, what it currently does and doesn't do, why major companies are still implementing it, and how to create one for your website in under 2 minutes using our free generator.
llms.txt is a plain-text file that lives at the root of your website — at yourwebsite.com/llms.txt — written in Markdown format. Its purpose: to tell AI language models which pages on your site contain your most important, accurate, and relevant content.
Think of it as a sitemap, but designed specifically for AI systems rather than search engine crawlers. Where sitemap.xml tells Google 'here are all my pages,' llms.txt says: 'If you're an AI system trying to understand what my business does, start here — these are the pages that matter most.'
It was proposed in September 2024 by Jeremy Howard, co-founder of Answer.AI (and fast.ai), as a community-led open standard. The initiative is hosted at llmstxt.org, and adoption has grown quickly — by October 2025, BuiltWith tracked over 844,000 websites with an llms.txt file in place.

These three files are often confused. Here's how they actually differ:

The key distinction: robots.txt is a directive — it can allow or block access. llms.txt is a guide — it signals priority, but it cannot force an AI system to read it or comply. Think of it as an invitation, not a gate.
A critical honest note: as of early 2026, no major AI platform has officially confirmed they actively use llms.txt to inform their responses. Google has stated its AI Overviews continue to rely on traditional SEO signals. OpenAI's GPTBot has been observed occasionally accessing the file, but infrequently. This is the current reality — and it's why we'll explain exactly why you should still implement it anyway.
Here's a real example of what an llms.txt file looks like for a digital agency like 12Grids:
# 12Grids
> 12Grids is a full-service CX and AI engineering company based in Mumbai, India.
> We design, develop, and market digital products for enterprises and government bodies.
## Core Services
- [UI/UX Design Services](https://www.12grids.com/services/ux-design): Our UX design process, approach, and case studies
- [Web Development](https://www.12grids.com/services/web-development): Custom websites, web apps, and platforms
- [Mobile App Development](https://www.12grids.com/services/mobile-app-development): React Native and Flutter apps
- [AI Integration](https://www.12grids.com/services/ai-development): AI products, chatbots, and intelligent systems
## Featured Work
- [Mumbai Zoo Case Study](https://www.12grids.com/case-studies/mumbai-zoo): Government digital ticketing system
- [SBI Payments](https://www.12grids.com/case-studies/sbi-payments): Banking UX transformation
- [UPL Global Platform](https://www.12grids.com/case-studies/upl): Multi-country CMS for Fortune 500 agro brand
## Useful Resources
- [Mobile App Cost Guide](https://www.12grids.com/articles/mobile-app-development-cost-india): Real cost breakdowns for 2026
- [AI Website Guide](https://www.12grids.com/articles/ai-website-development-india): What AI-enabled websites actually do
- [UX Audit Guide](https://www.12grids.com/articles/ux-audit-guide): Free checklist and professional service
The structure is simple: a short brand description at the top, followed by categorised sections linking to your most important pages. Each link includes a brief description explaining what the AI will find there.
The file is written in Markdown — the same format used in README files on GitHub or in documentation tools like Notion. No specialist technical knowledge is required to write one.
Given that no major AI platform has officially confirmed they read llms.txt, you might reasonably ask: why bother?
Here's our genuine assessment — built from 9 years of digital product work and 18 months of tracking AI search behaviour for our clients:
In 2003, only a small number of websites had an XML sitemap. By 2006, Google, Yahoo, and MSN jointly adopted it as a standard and it became essential infrastructure. llms.txt is following a similar adoption curve. The companies that implement it today — when it takes 10 minutes — won't be scrambling to implement it at scale when it becomes a recognised standard.
Anthropic (the company behind Claude, the AI you may be reading this through) has already published an llms.txt on their own documentation site. Cloudflare. Stripe. Zapier. These are not companies that waste engineering time on meaningless signals. They're building for where the web is going.
The process of creating an llms.txt file requires you to answer: which pages on my site are most important? Which content best represents our expertise and services? Which case studies are most compelling? This audit is genuinely valuable independent of whether AI systems read the file.
Most websites grow organically. Important pages get buried. Navigation becomes cluttered. Doing the llms.txt exercise forces clarity about what you actually want AI systems — and human visitors — to prioritise.
Even without llms.txt, AI systems are crawling Indian business websites. GPTBot, ClaudeBot, PerplexityBot — these crawlers are active. What they extract from your site is determined by how clearly your content is structured, how quickly the most important information is surfaced, and how much navigation noise they have to cut through.
An llms.txt file is one signal in this direction. A well-structured website with clear headers, direct answers, and comprehensive content is equally important — and llms.txt points AI systems to the pages where you've done that work well.
A robots.txt file that blocks the wrong crawler can seriously harm your SEO. A badly configured sitemap can cause indexing problems. llms.txt has no downside risk. If AI systems don't read it — you've lost nothing. If they do, or when they begin to, you're ahead. The implementation cost is 10-15 minutes. Our generator makes it 2 minutes.
Our honest position at 12Grids: llms.txt is not going to transform your AI visibility overnight. The companies claiming dramatic ranking improvements from llms.txt alone are overstating the evidence. But it's a legitimate, low-cost signal in an evolving landscape — and the companies who understand that are implementing it now, not waiting.
Because we believe in building trust through honesty, here's what llms.txt doesn't currently do:
Keeping this honest doesn't mean ignoring llms.txt. It means implementing it as one sensible layer of an AI visibility strategy — not as the whole strategy.
We built a free online tool specifically for this. Enter your website URL and key page details, and the generator creates a properly formatted, ready-to-upload llms.txt file for you.
Generate Your llms.txt File — Free
We built a free llms.txt generator tool specifically for Indian businesses. Enter your website URL, pick the pages you want AI systems to prioritise, and download a ready-to-upload llms.txt file in under 2 minutes. No developer needed.
→ Use the Free Generator: www.12grids.com/tools/llms-txt-generator
→ Want us to implement it properly along with a full AEO audit? Talk to us: sales@12grids.com
Open a plain text editor (VS Code, Notepad, or any text editor). Create a file named exactly: llms.txt (no variations). Use this structure:
# [Your Company Name]
> [One paragraph describing what your business does, who you serve, and your location]
## [Section Name — e.g., Services, Products, Resources]
- [Page Title](https://yoursite.com/page): Brief description of what's on this page
- [Page Title](https://yoursite.com/page): Brief description
## Optional: Full Content
For AI systems that prefer complete content:
- [Page Title](https://yoursite.com/page.md): Link to Markdown version of the page
Save the file as plain text (UTF-8 encoding). Upload it to your web server's root directory — the same folder that contains your robots.txt file. Verify it's accessible at: yourwebsite.com/llms.txt
Yoast SEO (version 22+) includes a built-in llms.txt generator with one-click activation. It automatically selects your most recent, high-quality pages, refreshes the file weekly via cron job, and lets you preview before publishing. If you're on WordPress and already using Yoast, this is the zero-effort path.
The pages you include in llms.txt should represent the best of what your website has to offer. Apply this selection framework:

A focused list of 15-25 genuinely strong pages is more useful to AI systems than an exhaustive list of 200 pages. Quality of selection signals that you understand your own content — which itself is an authority signal.
The llms.txt specification also mentions an optional llms-full.txt file. Where llms.txt provides a curated directory of links, llms-full.txt provides the actual full text content of those pages in a single Markdown-formatted file — making it easier for AI systems to read everything without fetching individual pages.
Think of it as: llms.txt = the index. llms-full.txt = the full content library.
For most Indian business websites, llms-full.txt is not necessary in 2026. It adds implementation complexity and file size management overhead without confirmed additional benefit. Our recommendation: implement llms.txt first and well. Add llms-full.txt only if you have a developer resource and content-heavy documentation that AI systems might genuinely benefit from reading in full.
The Indian digital market is moving faster toward AI-mediated discovery than most people realise. In our conversations with clients across e-commerce, B2B services, manufacturing, and financial services, we consistently see the same pattern: buyers who previously spent 20-30 minutes comparing websites via Google are now spending 5 minutes asking an AI assistant and then reaching out directly to 1-2 brands.
That behavioural shift has meaningful implications for who gets into the consideration set. And right now, the vast majority of Indian business websites are not optimised for AI discovery at any level — no AEO strategy, no structured content, no schema markup, and certainly no llms.txt.
The gap between the Indian businesses implementing AI visibility strategies today and those waiting represents a real first-mover window. It's the same window that existed with mobile SEO in 2012 and voice search in 2018. Early movers built advantages that compounded. Late adopters spent more to catch up.
For 12Grids, implementing llms.txt for our own website is part of a broader AEO strategy we've been building since early 2025. We use it alongside structured content, FAQ schema, entity optimisation, and citation building. The llms.txt file takes 10 minutes. The broader strategy takes months. Both matter.
Because EEAT requires experience — not just theory — here's what we've observed directly:
The honest summary: llms.txt alone will not change your AI visibility. A properly structured website, deep authoritative content, and consistent entity signals across the web will. llms.txt is the 10-minute foundation layer that supports the harder work — and there is no good reason not to have it in place.
Most llms.txt generators available online are built for international SaaS companies with developer audiences. They assume you know your way around a root directory and understand Markdown formatting.
We built ours for Indian businesses — marketing managers, business owners, and founders who want the right infrastructure in place without needing a developer.
What our generator does:
Generate Your llms.txt File — Free
We built a free llms.txt generator tool specifically for Indian businesses. Enter your website URL, pick the pages you want AI systems to prioritise, and download a ready-to-upload llms.txt file in under 2 minutes. No developer needed.
→ Use the Free Generator: www.12grids.com/tools/llms-txt-generator
→ Want us to implement it properly along with a full AEO audit? Talk to us: sales@12grids.com
After generating your file, if you want a full AEO audit — checking how your brand currently appears in AI search tools, identifying gaps in your entity definition, and building a prioritised action plan — we offer this as a standalone service. Reach out to sales@12grids.com and mention the llms.txt article.
No. Google has confirmed that its algorithms — including AI Overviews — rely on traditional SEO signals, not llms.txt. Implementing it will not improve or harm your Google rankings.
Not directly or immediately. Whether ChatGPT mentions your brand depends on how well your content answers relevant questions, how authoritative your site is, and whether your brand is known well enough to be in the training data or real-time retrieval results. llms.txt is a supporting signal, not a citation guarantee.
No. Blocking AI crawler access is done via robots.txt using User-agent rules for specific bots (GPTBot for OpenAI, ClaudeBot for Anthropic, PerplexityBot for Perplexity, etc.). llms.txt cannot restrict access — it can only guide it.
Update it when you publish significant new content (new pillar articles, major case studies), when you launch new services, or when important pages change URLs. Quarterly reviews are a reasonable cadence for most businesses.
Not yet. It's a community proposal with growing adoption. It has not been formally adopted by W3C, IETF, or any standards body as of early 2026. Major companies are implementing it voluntarily because they believe in where it's headed — not because it's required.
Should I create separate llms.txt files for different languages?
If your website has separate language sections (e.g., Hindi and English), you can create language-specific llms.txt content by including both language versions of your pages. A single llms.txt file at the root can reference pages from any section of your site.
How is llms.txt different from AEO (Answer Engine Optimisation)?
llms.txt is one small tactical element within the broader discipline of AEO. A complete AEO strategy includes: content depth and structure, FAQ schema implementation, entity optimisation across all digital touchpoints, citation building in authoritative sources, and consistent brand definition. llms.txt takes 10 minutes. A comprehensive AEO strategy takes months. Both are worth doing.
llms.txt is a small file with a big purpose. Right now, it's a signal in an evolving landscape — not a definitive ranking factor. The companies that will benefit most from it are the ones implementing it as part of a broader commitment to AI-readable content and AEO strategy, not the ones treating it as a magic fix.
For Indian businesses in 2026, the question isn't whether AI-mediated discovery is coming. It's already here. The question is whether your website is structurally prepared for how AI systems read, understand, and cite your brand when a potential customer asks the right question.
Your llms.txt file is the first 10 minutes of that preparation. Generate yours now using our free tool. Then come talk to us about the rest.
We will be launching soon a free llms.txt generator built specifically for Indian businesses.
It will let you enter your website URL, choose the pages you want AI systems to prioritize, and download a ready-to-upload file in under 2 minutes, no developer needed.
→ Want us to implement it properly along with a full AEO audit? Talk to us: sales@12grids.com

_desktop_list_webp_e848d9a5.webp)
5 mins : 09 Mar 2026

5 mins : 11 Nov 2025

7 mins : 03 Nov 2025