If you run a small business in Philly and you’ve been hearing about llms.txt from an agency pitch or a LinkedIn post, you’re probably wondering if it’s the next thing you have to do before AI search leaves you behind. The honest answer is: probably not yet, and definitely not before a handful of other things. Here’s what the file actually does, what’s been confirmed about it, and where it really sits on your priority list.
What llms.txt Actually Is
llms.txt is a plain-text or Markdown file you place in your website’s root directory, the same location as robots.txt. It’s modeled on that same idea: a machine-readable signal file sitting at yoursite.com/llms.txt. But instead of talking to search engine crawlers, it’s meant to talk to large language model crawlers, the bots that (in theory) power tools like ChatGPT, Claude, and Perplexity.
The pitch behind it is simple and, on paper, appealing. You write a curated list of your most important pages, a short description of who you are and what you do, and sometimes a note about whether you want your content used for AI training. The theory is that an AI crawler reads this file first, understands your site faster and more accurately than it would by crawling everything, and cites you more reliably as a result.
What llms.txt Promises
The promise, in short, is control and clarity. A well-built llms.txt gives an AI system:
- A short, direct description of your business and what it does
- A curated list of your most important pages, ranked by relevance
- Optional directives about whether your content can be used for AI model training
That’s a reasonable thing to want. Every business owner would like more accurate representation in ChatGPT or an AI Overview. The problem isn’t the idea. It’s the evidence that anyone is actually reading the file.
The Honest 2026 Evidence
This is the part most of the content ranking for this question skips over or buries in a footnote. Google’s John Mueller has publicly compared llms.txt to the old keywords meta tag, a signal file businesses filled out in good faith that search engines simply stopped using. He’s also said that server log data shows AI crawlers do not even request the llms.txt file when they visit a site. Not that they request it and ignore it. That they don’t ask for it at all.
As of mid-2026, no major AI provider, not OpenAI, not Google, not Anthropic, not Perplexity, has publicly confirmed that its crawlers read or use llms.txt in any way. That’s a meaningful gap between the marketing around this file and what’s actually been verified. A lot of the content pushing llms.txt implementation checklists comes from SaaS tools and agencies selling AEO or GEO services, and they have an incentive to make the file sound more consequential than the evidence currently supports.
Why “No Confirmed Adoption” Doesn’t Mean Zero Risk to Try
None of this means llms.txt is a scam or that you should actively avoid it. It means you should be honest about the cost and the realistic upside before you spend developer time on it. A minimal llms.txt file takes very little effort to build. It won’t hurt your site, it won’t confuse Google’s regular search crawler, and if any AI provider does start reading it later, you’ll already have one in place.
What it won’t do is move the needle on whether ChatGPT recommends you, whether you show up in a Google AI Overview, or whether Perplexity cites your business. If an agency is quoting you real hours and real dollars to build an elaborate llms.txt strategy, that’s money better spent elsewhere first. If your developer has an idle hour, there’s no harm in it. That’s the honest distinction.
The Real Priority Order for AI Search Visibility in 2026
If you’re trying to figure out where to actually spend time and budget to show up in ChatGPT, Google AI Overviews, Perplexity, or Claude, here’s the order that reflects how these systems actually pull and trust information, from highest leverage to lowest:
- Domain authority and trust signals. Reviews, backlinks, consistent citations across the web, and a Google Business Profile that’s verified and active. This is the foundation everything else sits on.
- GEO-ready content structure. Clear, direct answers to the questions your customers actually ask, written in plain language an AI model can lift and quote accurately without misrepresenting you.
- Schema markup. Structured data that tells search engines and AI systems exactly what your business is, where it’s located, and what it offers, in a format machines parse reliably.
- llms.txt. Hygiene, not strategy. Worth having if it’s free to add, not worth prioritizing over anything above it.
Notice what’s missing from the list of things that actually drive AI visibility: a signal file that crawlers apparently aren’t even requesting. If you want the full picture of how Philly businesses actually get surfaced in AI tools, our guide on how Philadelphia businesses show up in ChatGPT and AI search walks through the mechanics in more depth.
How to Build a Minimal, Honest llms.txt in Under an Hour
If you’ve got the developer time free and want to have one in place just in case adoption changes, here’s how to do it without overbuilding:
- Create a plain text file named llms.txt and place it in your site’s root directory, so it lives at yoursite.com/llms.txt
- Start with a one or two sentence description of your business: who you are, where you’re located, what you do
- List your 5 to 10 most important pages with a short line describing each one, in plain Markdown format
- Skip the elaborate categorization schemes some tools push. A short, accurate file beats a long, over-engineered one
- Keep it updated the same way you’d update a sitemap, as an occasional maintenance task, not a recurring project
That’s it. There’s no reason to pay for a specialized tool or a multi-hour consulting engagement to produce this file.
What Philly Small Businesses Should Do Instead, First
Before you touch llms.txt, put your time into the things with confirmed, meaningful impact on how you show up in AI-generated answers and regular search:
- Get your Google Business Profile fully optimized, verified, and consistently updated. Our Google Business Profile optimization guide covers the specifics for Philly businesses.
- Build up review volume and respond to reviews consistently. This is one of the strongest trust signals both Google and AI models draw on.
- Add or clean up schema markup so your business details are structured and unambiguous.
- Write clear, direct on-page answers to the questions your customers actually search for, rather than vague marketing copy.
- Be careful about how you produce that content. If you’re leaning on AI tools to write it, read our guide on whether AI-generated content is bad for SEO before you scale it up.
None of this happens overnight, and that’s worth knowing going in. If you want a realistic sense of timelines, our breakdown of how long SEO actually takes to work in Philadelphia sets expectations without the hype.
How Media Proper Approaches AI-Search Readiness
We’re not going to sell you an llms.txt package, because the evidence doesn’t support charging for one. Our Philadelphia SEO services focus on the things that actually move the needle for AI and traditional search alike: authority, structured data, clear content, and a Google Business Profile that’s doing its job. When the evidence changes on tools like llms.txt, we’ll adjust. Until then, we’d rather spend your budget on what’s confirmed to work than on hygiene dressed up as strategy.