What is llms.txt?
A curated text file that maps your key content for AI.
llms.txt is an emerging convention: a plain-text file at /llms.txt that points AI systems at your most important pages, like a hand-drawn map. It is a sibling to robots.txt, but descriptive rather than restrictive.
Set expectations honestly, adoption is low and there is little evidence major engines use it for citation yet, though it helps agentic/developer tooling. Ship it as cheap hygiene, not a silver bullet; the full take (with the ROI caveats) is in what is llms.txt. CiteTrack generates and serves it, kept consistent with robots.txt.
Why — and when — does it matter?
Use llms.txt as low-cost content discovery hygiene for systems that support it. It is not an access-control file and does not guarantee indexing or citation.
How it works
Publish a plain-text file at the site root.
Describe the site and link canonical high-value resources.
Keep URLs current and accessible.
Validate format and server response.
Measure downstream crawling separately.
Worked example
A documentation site lists its API overview, authentication guide and changelog instead of dumping every URL, giving compatible agents a curated starting map.
How to track and report it
Reporting rule: State the engine, prompt set, market, time window and denominator. A number without scope cannot be reproduced or compared.
Practical playbook
- Publish a plain-text file at the site root.
- Describe the site and link canonical high-value resources.
- Keep URLs current and accessible.
- Validate format and server response.
- Measure downstream crawling separately.
Common mistakes
- Using it to block crawlers
- Claiming guaranteed ranking benefits
- Publishing a stale auto-generated URL dump
Questions people ask
What is llms.txt?
A curated text file that maps your key content for AI. Use llms.txt as low-cost content discovery hygiene for systems that support it. It is not an access-control file and does not guarantee indexing or citation.
How does llms.txt work?
Publish a plain-text file at the site root. Describe the site and link canonical high-value resources. Keep URLs current and accessible. Validate format and server response. Measure downstream crawling separately.
How do you measure llms.txt?
Valid response and current canonical URLs. Observed requests to listed pages. No credible citation-lift formula exists; treat adoption and impact separately.
What is a common mistake with llms.txt?
Using it to block crawlers. The scope and methodology should be stated before interpreting the result.
llms.txt Generator
Use the free tool to turn this reference into a concrete check on your own site.
Open the tool →