Google Addresses llms-author.txt and Online Identification

Search Engine Journal reported that Google's John Mueller responded to a Reddit user who was testing an llms-author.txt file and Content-Signal directives to improve online identification for a name that overlaps with more prominent entities. The report says Mueller stated that Google does not use llms.txt or llms-author.txt, and that he did not know of other crawlers or LLMs confirming use of those files beyond SEO tools.

The distinction matters because several ideas are being conflated: the broader /llms.txt proposal, a non-standard llms-author.txt variant, Cloudflare's Content Signals policy, and Cloudflare's Markdown for Agents feature. Cloudflare documents Content Signals as machine-readable robots.txt directives for search, AI input, and AI training preferences, while Markdown for Agents converts eligible HTML pages to Markdown when clients request text/markdown. That does not mean Google Search treats those signals as ranking or identity inputs.

Use these files only as supplemental documentation when they match a real crawler or agent workflow. For discoverability and entity disambiguation, the safer work is still durable web evidence: author pages, organization profiles, schema, canonical links, bylines, citations, and third-party references that search systems can corroborate.

The meaningful signals are explicit support announcements from major crawlers, measurable server-log fetches from known AI agents, and documentation that explains how a platform consumes the file. Without those signals, llms-author.txt should be treated as an experiment rather than an SEO control.