Listes d’alerte aux arnaques : qui les tient et quand leur faire confiance
Before anyone loses money on an onion service, they usually check a scam list first. That instinct is sound. But the lists themselves have maintainers, incentives, and failure modes that most readers never stop to examine.
Who Actually Maintains These Lists
There is no central registry of darknet fraudsters. Scam warnings come from a patchwork of hobbyist maintainers, forum moderators, security researchers, and commercial directories, each with different standards for what counts as proof. The best-known example from the underground itself is Ripper.cc, launched in 2016 as a kind of Yelp for cybercriminals, where users submitted evidence-backed complaints about "rippers" who defrauded their own peers. Motherboard reported that its creators tried to differentiate the site by publishing their plugin source code on GitHub and recruiting forum admins as moderators.How Verification Is Supposed to Work
The strongest verification model in this space is cryptographic rather than editorial. Dark.fail pioneered the Onion Mirror Guidelines, which require every site to publish its PGP keys at /pgp.txt and a signed mirror list at /mirrors.txt, so readers can confirm addresses themselves instead of trusting a directory's word. The published specification also mandates a canary file re-signed every fourteen days, proving the operator still controls the key. Sites that refused to comply were demoted and marked unverified. The lesson generalizes: a good list shows its work, it does not merely assert conclusions. Automated blocklists take a different approach. Jarelllama's Scam Blocklist, for instance, filters candidate domains against whitelists and the Tranco ranking, then routes anything ambiguous to manual review before inclusion. Human judgment remains the bottleneck by design.The Paid Listing Problem
Money changes everything. According to reporting by BleepingComputer, the older Russian-language blacklist Kidala.info was widely suspected of removing scammers from its lists after they paid a fee, which is precisely why Ripper.cc made non-negotiable listings its selling point. The same dynamic applies in reverse. A directory that accepts advertising from the markets it ranks cannot credibly rank them. Any list whose business model depends on the entities it judges deserves extra skepticism, not automatic trust. Readers rarely see these incentives directly. Sponsorship banners get framed as support for uptime costs, and featured placements get justified as quality signals. The honest directories disclose their funding model up front; the rest leave readers to guess.False Accusations Are Cheap
Rick Holland of Digital Shadows put the core problem plainly when Ripper.cc launched: if fake Yelp reviews exist, fake ripper reports will too. Nothing stops a competitor from fabricating chat logs and filing a convincing complaint against an honest vendor. Communities fight back with process, not perfection. In 2020, Dread's moderators banned the BitBazaar subdread after identifying 454 sock-puppet accounts used to inflate subscriber numbers, according to Digital Shadows analysts. The manipulation was caught because it left patterns. But even well-run forums stumble. The Tor Times documented a year-long controversy in which vendors accused Dread administrator HugBunter of operating shop-hosting under a second alias and vanishing with payments, allegations he denies and which remain unresolved. When the accuser and the accused are anonymous, reputation is all there is.How a Careful Reader Should Use Them
A scam list is a starting point for investigation, never a verdict. Treat it the way an analyst treats any single-source claim:- Check whether the list explains its evidence and lets you review the underlying report.
- Look for conflicts of interest: paid placements, affiliate links, or listings that double as advertisements.
- Cross-reference across independent sources before acting on any single accusation.
- Prefer cryptographically verifiable claims over editorial assertions whenever possible.
A warning list tells you what someone else believes. Only verification tells you what is true.The readers who stay safe are the ones who treat community warnings as leads, verify them independently, and remember that today's trusted curator can be tomorrow's headline. Our security notes cover more verification techniques, and our guide to link directory sites explains how PGP-based verification became the standard.