Google’s John Mueller answered a question about something a Redditor was using called llms-author.txt.
The person who posted the question is having trouble finding their name because it is the same as two other, more popular entities, making it harder for people to find them and learn about their services.
The question asked:
“For years, this meant that part of my “who is this person” signal was diluted or downright wrong every time a model or search function tried to sum me up.
FIRST STEP
First, an explicit llms-author.txt file, separate from the main llms.txt file, stating the job title, agency, location, and practice area in simple sentences rather than relying solely on an outline to carry this weight.
SECOND STEP
Second, add the new Content-Signal header (ai-train=no, search=yes, ai-input=yes) to robots.txt, mainly out of curiosity as to whether the header-level intent statement changes anything measurable.
What they wanted to know was if anyone had done controlled testing on the Content-Signal headers to see if they worked.
LLMs-author.txt
Apparently the LLMs-author.txt may not be a real thing, as there is no official proposal for it, much less a real standard for it.
Content signal header
It’s a little more confusing. Cloudflare had proposed a Content-Signal directive for robots.txt. Cloudflare then used the same Content-Signal syntax as an auto-generated HTTP response header in Markdown for Agents. Markdown for Agents is a Cloudflare feature that automatically offers a Markdown version of a web page whenever a web client requests it.
So, of the two items used by the Redditor, neither appears to be part of an actual proposal.
Google’s take on LLMs-Author.txt
Google’s John Mueller correctly called Cloudflare’s Content-Signal a robots directive, which is formally the proposal.
He replied:
“I guess some things…
* Google does not use llms.txt or llms-author.txt. I don’t know of any other bots/llms confirming that they use them (other than SEO tools).
*AFAIK, none of the crawlers/llms use robots.txt “content-signal” directives. It was made by a CDN, as far as I know it has no effect for a robot or llm. Using it just adds bloat and future maintenance to your robots.txt file.
You can also add other arbitrary elements to your robots.txt file, the robots just use the directives they support and ignore the rest.
Redditor’s Solution May Not Be Technical SEO
The Redditor is looking for a technical SEO solution to solve a problem that goes well beyond technical SEO. LLMs, chatbots and search engines rely on signals from the web, including structured data. But in this situation, since the web indicates that these other entities have the same name, the Redditor’s problem is really with the web and not a technical SEO problem to solve. They can find a satisfactory solution to their problem by making themselves better known on the web, perhaps by being interviewed on videos, podcasts and on web pages. Maybe they should do more notable things to expand their digital footprint on the Internet.
Featured image by Shutterstock/Ken Wolter
