Dev.to WebDev 🛠 Dev 👁 0 📖 4 min read

Adding llms.txt to a Law Firm Site: What It Can and Cannot Do

Sooner or later, a law firm client forwards an article saying every website needs an llms.txt file to show up in ChatGPT. The request lands on whoever maintains the site, usually with a deadline attached. For developers

Sooner or later, a law firm client forwards an article saying every website needs an llms.txt file to show up in ChatGPT. The request lands on whoever maintains the site, usually with a deadline attached. For developers who build criminal defense firm sites, it is worth understanding what the file actually is before deciding how much time it deserves.

The short version: llms.txt is a proposal, and no major search engine has committed to using it. Adding one is cheap and harmless. Treating it as a way to get a firm named in AI answers is a mistake, and the time is usually better spent on things AI crawlers and search engines do use.

What llms.txt is

The llms.txt idea was proposed in September 2024 by Jeremy Howard of Answer.AI. The file sits at the root of a site, like robots.txt, but it is written in Markdown. It gives a short summary of the site and a curated list of links to the pages a language model should read, sometimes pointing to plain Markdown versions of those pages.

The original motivation was documentation sites, where a model trying to answer a coding question benefits from a clean index of the docs. A law firm site is a different case. It has a handful of practice pages, attorney bios and a blog, all of which are already linked from normal HTML navigation.

What Google and others have said

Google has said that llms.txt files will not help or hurt a site's rankings in Google Search. John Mueller of Google has also been openly skeptical that AI systems rely on the file at all. Most AI companies have not said publicly whether their crawlers read it, so any claim that a specific assistant depends on it deserves a source.

None of that means the file is harmful. It means a firm should not expect the file to change whether ChatGPT or Gemini mention the practice. Those tools draw on what they can crawl and what other sources say about the firm, and a short index file at the root does little to change either.

If the client still wants one

A minimal file takes a few minutes and keeps the client happy. Keep it accurate and short. For a defense firm, a reasonable version looks like this:

# Smith Defense Law

> Criminal defense firm in Sacramento County, California, handling DUI, drug and domestic violence cases.

## Practice areas
- [DUI defense](https://example.com/dui/): DUI charges and DMV hearings
- [Drug charges](https://example.com/drug-charges/): possession and sales cases

## About
- [Attorneys](https://example.com/attorneys/): bios and bar admissions

Treat the descriptions as public advertising copy. They should match the website exactly and follow the state bar's advertising rules, so avoid words like "expert" or claims about results. If the firm changes its practice areas, the file needs updating too, which is one more thing to forget.

Where the effort pays off more

The things that help AI tools understand a defense firm are the same things that help search engines. Start with access. Check robots.txt and any firewall or CDN bot settings to confirm the AI crawlers the firm wants to allow can reach the site. This guide on which AI crawlers a criminal defense website should let in goes through the main user agents and the trade-offs.

Next, make sure the important content is in the server-rendered HTML. Many AI crawlers do not execute JavaScript, so a charge page that loads its text client-side may look empty to them. Then make each page clear about who the firm is, where it practices and which charges it handles, in plain text near the top. Consistent Organization and LegalService structured data, matching the visible content, helps machines connect the details.

Finally, keep facts consistent across the firm's site, Google Business Profile and directory listings. AI answers often combine several sources, and conflicting addresses or practice areas are a common reason they describe a firm wrongly.

Measuring before and after

If you add llms.txt, log requests to it. Most hosting platforms and CDNs let you filter access logs by path and user agent. After a few months, you will have your own evidence of whether anything reads it. Compare that with requests from known AI crawlers to the rest of the site, which tells you far more about how the firm's pages are being used.

More on AI search and legal marketing technology is collected in the library on ChatGPT, AI search and legal marketing technology.

Questions about llms.txt for law firms

Will llms.txt get a defense firm mentioned in ChatGPT?

There is no public evidence that it does. Be cautious with anyone who promises otherwise.

Can llms.txt hurt a site's rankings?

Google has said it will not help or hurt rankings in Google Search. The main risk is outdated or inaccurate content in the file.

Should the file list every page on the site?

No. If you add one, keep it to the main practice pages and attorney information, and update it whenever those change.

SBD Marketing works only with criminal defense law firms, one firm per city. It builds websites for defense practices, handles SEO for the charges they defend, works on visibility in AI search tools such as ChatGPT, Gemini, Claude and Perplexity, and runs SEO audits.

📰 Read the original article on Dev.to WebDev

Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.