Actuent Bot
The crawler behind Actuent, a search engine for AI agents
If you've seen this in your logs, it was Actuent:
Mozilla/5.0 (compatible; Actuent/1.0; +https://docs.actuent.ai/bot)Actuent reads public websites and turns them into LAWP, a structured summary of a site's pages and the actions visitors can take, so AI assistants can understand sites without scraping raw HTML.
What it fetches
/robots.txt, which it always reads first./.well-known/lawp.jsonif you publish one. When it exists, Actuent uses it instead of crawling.- Your homepage, and for popular sites
/pricing,/aboutand/contact. - A page someone searches for directly, such as
yoursite.com/shoes.
It only reads public pages: no logins, no forms, and it doesn't follow every link.
robots.txt
Actuent follows robots.txt (RFC 9309). It uses rules for Actuent if you have any, otherwise your * rules. To keep Actuent out completely:
User-agent: Actuent
Disallow: /To block only part of your site:
User-agent: Actuent
Disallow: /account/Changes are picked up within about an hour. If your robots.txt returns a server error, Actuent treats the site as off-limits until it's reachable again.
Some fetches go through Jina Reader
Some pages are fetched through Jina Reader so JavaScript-heavy sites can be read. Actuent checks your robots.txt before asking Jina for a page.
Take control instead of being crawled
Publish your own LAWP at /.well-known/lawp.json. You decide exactly what AI agents see, you rank higher in Actuent, and you can make actions like bookings executable. See LAWP Actions.
Contact
Questions, or want your site's data corrected or removed? Email support@localilabs.com. Actuent is made by localilabs.