HottubBot

HottubBot is the web crawler of Search by Hottub, run by Hottub (joinhottub.com). It fetches public pages: mostly site home pages, websites of businesses in the Hottub Directory, and pages other sites link to.

Search only

What it fetches serves search results that link to your page with its title and a short excerpt. It is never used to train AI models, never sold or licensed for training, and never turned into AI-written summaries. Pages do feed Search by Hottub's own ranking, which only orders results.

How to recognize it

User-Agent: Mozilla/5.0 (compatible; HottubBot/1.0; +https://hottub.tech/bot)
robots.txt name: HottubBot

HottubBot connects only from these addresses (also in /bot.json):

35.170.128.247/32
54.166.142.222/32

44.240.166.76/32: used only for check it now. When someone using Search by Hottub asks whether a result is still current, Hottub fetches that single page once from this address, as HottubSearch/1.0 (+https://hottub.tech; feedback@joinhottub.com), following the same robots.txt rules. It never crawls.

How it crawls

It reads robots.txt first (kept up to 24 hours) and follows the group for HottubBot, or * when there is none: Allow, Disallow and Crawl-delay.
One request at a time per site, at most one per second, slower if Crawl-delay asks.
On 429 or 503 it waits as Retry-After says; asked repeatedly, it leaves the site for that crawl.
HTML only, up to 2 MB and 5 redirects, with conditional requests: an unchanged page costs almost nothing.
No logins, forms, cookies or paywall workarounds, ever.

AI preferences and licences

search=n in a Content-Usage header or robots.txt line (the IETF AI-preferences draft), Cloudflare's Content-Signal: search=no, or an RSL licence that does not allow search for free keeps a page out. train-ai changes nothing: HottubBot never trains.

Limit or block HottubBot

Add rules to /robots.txt:

# Keep HottubBot out of the whole site
User-agent: HottubBot
Disallow: /

# Slow it down: seconds between requests
User-agent: HottubBot
Crawl-delay: 10

For a single page, use <meta name="robots" content="noindex"> (or name="hottubbot" for HottubBot only), or X-Robots-Tag: noindex. HottubBot also honors nofollow, nosnippet and data-nosnippet.

Removal and contact

A page marked noindex, nosnippet or search=n, a page that is gone, or one robots.txt closes leaves Search by Hottub at the next index update, with any copy it holds from Common Crawl. If robots.txt closes your whole site, the whole site leaves.

Questions, problems or faster removal (send the URLs): feedback@joinhottub.com. Machine-readable details: /bot.json.