RuleRadarBot

What it does

RuleRadar is a compliance scanner for food supplement listings. When someone pastes a product URL into it, our server fetches that one page, reads the product title and description, and checks the wording against EU rules on health claims.

Then it forgets the page. We do not store the content, we do not index it, and we do not use it to train anything.

What it does not do

• It does not crawl. One request, one page, one human who clicked a button. It never follows links to find more pages.
• It is not scheduled. There is no queue and no revisit schedule. If nobody asks, we never come.
• It does not scrape prices or stock. We read the product text. Prices, stock levels and buy buttons are filtered out before the check even runs.
• It respects robots.txt. If your rules disallow the path, we stop and tell the user to paste the text instead.

Rate and load

One request per scan, triggered by a person. A single visitor is limited to a handful of scans per day. In practice this means your server sees us at roughly the rate of a curious customer opening product pages — not a crawler.

We time out after 12 seconds and stop reading after 3 MB, so a slow or large page costs you one brief connection at most.

How to allow it

Most blocks we see come from generic bot protection that rejects any unfamiliar user agent. Allowing us is a user-agent rule, not an IP rule — we run on shared cloud infrastructure and our addresses change, so please do not allowlist by IP.

Cloudflare

Security → WAF → Custom rules. Add a rule with the expression (http.user_agent contains "RuleRadarBot") and the action Skip — selecting the managed rules and bot fight mode you want it to bypass. Place it above your blocking rules.

robots.txt

If you use robots.txt to control access, this allows us explicitly while your other rules stay as they are:

User-agent: RuleRadarBot
Allow: /

Other firewalls and WAFs

Any rule matching the user-agent string RuleRadarBot works. We do not disguise ourselves as a browser, so the string is stable and easy to match.

If you would rather we stayed away

Disallow us in robots.txt and we will stop, with no argument and no workaround:

User-agent: RuleRadarBot
Disallow: /

People can still scan your product text by pasting it in by hand, the same way they could paste it into any other tool. That is outside what a robots file governs, and outside what we can control.

Contact

Questions, or something behaving badly? Write to hello@ruleradar.eu and a person will answer.