Updated July 29, 2026 · Proposal status checked against llmstxt.org
What Is llms.txt? Proposal, Format & Limits
The Current Status
The project's own documentation calls llms.txt a proposal to standardize how websites make selected information available at inference time. It also says the proposal does not specify how applications process the file. That distinction matters: a public format can be useful without being an adopted standard.
What you can safely claim
You can publish an llms.txt file in the proposed format. You cannot infer from its existence that a particular crawler reads it, that content is indexed, or that a page will receive more AI citations.
Google's July 2026 generative Search guidance is explicit: Google Search does not use llms.txt, and publishing one neither helps nor harms visibility or rankings in Google Search. A test therefore needs to name the other application it is intended for.
Proposed File Format
The proposal uses Markdown: an H1 for the project or site, an optional blockquote summary, explanatory text, and H2 sections containing lists of links. The example below is intentionally small so it can be maintained.
# Example Company
> Official product documentation and support resources.
## Documentation
- [Getting started](https://example.com/docs/getting-started): Core setup
- [API reference](https://example.com/docs/api): Current API documentation
## Optional
- [Changelog](https://example.com/changelog): Product changesUse absolute, canonical URLs and descriptions that remain true outside the surrounding page. If you also publish Markdown versions, keep them synchronized and avoid contradicting the canonical HTML.
llms.txt vs robots.txt vs sitemap.xml
| File | Role | Status |
|---|---|---|
| robots.txt | Crawler access directives for compliant crawlers | Established crawler mechanism |
| sitemap.xml | URL discovery and update hints | Supported by major search engines |
| llms.txt | Optional curated context in the proposed format | Proposal; application behavior is unspecified |
Keep robots.txt, sitemaps, crawlable links, canonical HTML, and normal indexing controls correct regardless of whether you experiment with llms.txt.
A Responsible Implementation Test
- Choose a small set of current, authoritative pages.
- Publish a valid UTF-8 text file at the site root using the proposed syntax.
- Do not expose private URLs, credentials, drafts, or pages blocked for a reason.
- Keep the listed URLs available through normal crawlable links and sitemaps.
- Check server logs for requests to /llms.txt and record the requesting user agent.
- Measure citations or referrals separately; do not attribute a change without a controlled baseline.
- Remove or update stale entries when the linked source changes.
Primary Sources
Continue Learning
Frequently Asked Questions
Ready to Scale Your SEO?
Generate optimized content, review it with SEO checks, and publish to WordPress from one workflow.
Start 3-Day Free Trial