SlopGuard
SlopGuard reads your live pages the way a first visitor would and quotes what they’d trip on. Four AI reviewers go over the pages; an adversarial pass tries to kill every finding against the quoted text. What survives is a markdown report with paths, line numbers, and a fix scope, ready to paste into a coding agent.
Sample report
We ran SlopGuard on this site before it shipped. The receipts document the pages as they stood then. Every published run so far is of Hexento’s own pages; queue slot 002 will be the first external one.
0 trust-killers
6 embarrassing
16 nits
SG-006
embarrassing
self-scored metric
1/4 lenses
The 204/204 receipt stat is guaranteed by construction, yet is sold as an audit result.
The same page says any finding whose receipt cannot be re-found is dropped before the report is written, so the published set will always score N/N.
claim — method/index.html:72 (as the page stood, 2026-08-09)
Receipts are re-verified mechanically before publication; in run 001, 204 of 204 receipts passed that audit.
The exact kind of skeptical reader this site courts will notice the metric cannot fail.
Open the full report. Copy-for-agent is on that page. An archived report of the previous version of this site is public (158 verified findings).
What it costs
Your first report is free and complete. Same-site retries are free if the scan fails. If a complete report for that domain is still in the 24-hour window, you get that report back rather than a new run. A different site, or the same site after its window lapses, is free too. Paid features (scheduled reruns, diffs, a CI gate) are not for sale yet. No price to quote until they work.
How a finding is made
Your pages are crawled and rendered in a real browser, checked deterministically, then read by four independent AI lenses. An adversarial pass re-checks every finding against the quoted page. The report comes out as markdown. Method is the full pipeline. The optional roast queue is how an unlisted report becomes public, later, with your OK.