Guide
llms.txt for Regulated Product Websites
llms.txt is a lightweight, machine-readable overview of a website intended for large language models. This guide covers how regulated product companies can publish one responsibly, what to include, and what to avoid.
Last updated: June 2026
Definition
llms.txt
A proposed, plain-text file at the root of a website that provides a curated summary of the site and links to the sources an operator wants AI engines to consult when answering questions about the organization or its products.
Who this guide is for
- SEO and content teams working on AI visibility
- Regulatory Affairs and Quality Assurance
- Medical and clinical affairs
- Web engineering and technical SEO
- Product marketing
- Compliance and legal reviewers
What an llms.txt should cover for regulated products
Preferred one-line summary
A concise, review-approved statement of what the organization does.
Official page links
Links to product pages, IFUs, support content, and regulatory pages.
Scope statements
Clear language about what the company does and does not do.
Excluded content
Explicit exclusion of unapproved claims and off-label content.
Regional and language notes
How regional differences are represented.
Change control notes
Owner, review process, and update cadence for the file.
Example prompts
Illustrative prompts from a typical scoping exercise. Actual prompt libraries are tailored to your product portfolio, risk categories, and regions.
- Prompt
What does [Company] do?
- Prompt
Where can I find the official IFU for [Product]?
- Prompt
Which regulatory agencies has [Company] engaged with?
- Prompt
What products does [Company] make?
- Prompt
How do I contact [Company] for support?
- Prompt
Which markets does [Company] serve?
Example findings
Illustrative finding rows. Each finding includes the prompt, channel tested, observed issue, a risk rating, and a recommended action.
| Prompt tested | Channel tested | Observed issue | Risk level | Recommended action |
|---|---|---|---|---|
| What does [Company] do? | ChatGPT | Summary sourced from a third-party profile instead of official copy. | Medium | Publish llms.txt with review-approved one-line summary; link to official About page. |
| Where can I find the official IFU? | Perplexity | Engine cites a reseller-hosted IFU copy. | High | Publish llms.txt with direct IFU links; strengthen citation authority. |
| Which markets does [Company] serve? | Google AI Overview | Regional coverage stated inaccurately. | Medium | List regional pages in llms.txt with hreflang alignment. |
| How do I contact [Company] for support? | Brand Chatbot | Bot returns outdated support URL. | Low | Update llms.txt and knowledge base with current support URL. |
Illustrative examples.
Deliverables
Each engagement produces a structured evidence package designed to be reviewed, prioritized, and acted on.
- llms.txt draft aligned to review-approved copy
- Official source link inventory
- Scope and exclusion statements
- Regional and language guidance
- Change control and owner assignment
- Post-publish AI answer verification pass
Frequently asked questions
What is llms.txt?
llms.txt is a proposed convention that provides a curated, machine-readable overview of a website for large language models. It is not a standard and is not required, but a growing number of AI tools consume it.
Is llms.txt a regulatory record?
No. It is a convenience file for AI engines. It does not replace controlled labeling, IFUs, or approved claims.
What should regulated companies include?
Preferred one-line summary, links to official product pages, links to IFUs and support content, and clear scope statements about what the company does and does not do.
What should regulated companies avoid including?
Do not include unapproved claims, off-label suggestions, sales language, or any content that has not been through your normal review process.
How does llms.txt relate to citation authority?
llms.txt is one signal among many. It helps AI engines discover official sources, but strong citation authority still depends on crawlable HTML, structured data, and trustworthy source hierarchy.
Ready to see what AI is saying about your products?
Request a scoped AI Answer Audit for your product portfolio and risk categories.