‘Open and helpful community’ – of credit card thieves

Honeynet shows carders are getting slick

By Kevin Poulsen, 12th July 2003

Regcast training : Hyper-V 3.0, VM high availability and disaster recovery

Credit card fraud "power users" with programming skills and no fear are making it easier for newbies to break into white collar crime, according to a report from the Honeynet Research Alliance this week.

The report draws on data gathered earlier this year when a fraudster looking for a random host to put between himself and IRC wound up cracking a research honeypot maintained by students and faculty at Azusa Pacific University, as part of a loosely affiliated gaggle of deliberately hackable hosts and networks organized around the non-profit Honeynet Project.

Researchers secretly monitored the intruder as he joined an IRC channel on DALnet dedicated to obtaining, verifying and swapping credit card numbers, along with matching names, addresses, and everything else a good carder needs to begin ordering goods and services illicitly.

From early April to mid-May they watched the intruder move through a dozen chat rooms with names like "#ccinfo," "#ccpower," and "#virgincc." They also joined some of the channels themselves. They found a surprisingly open and helpful community of credit card thieves, where experienced fraudsters offered advice to newcomers, and stolen credit cards were given away freely to neophytes -- at least, in small amounts.

"They weren't trying to hide this at all, it was just completely out in the open," says Patrick McCarty, an undergraduate at the university, and a co-author of the report. "You'd think they would want to keep a lower profile."

Carding Commands

The researchers were also impressed by the level of automation that a handful of sophisticated carders brought to the scene. Fraud-oriented IRC bots made the channels more than just a communications medium. Carders could type in commands like "!chk" to verify that a credit card number is correct, and "!bank" to identify the bank that issued a particular card.

Daring fraudsters looking to get credit card numbers directly from a vulnerable e-commerce site could avail themselves of the "!cardable" command, which returns the URL for sites known to be vulnerable to attack. For more help, the "!exploit" command yielded URLs that a beginner could cut-and-paste into their browser to exploit known application-level Web server attacks. If they weren't up for cracking a host personally, the "!cc" command dispensed a single stolen credit card number from a database.

"Users need master only a series of custom IRC commands to carry out many key activities of credit card / identity theft," the report found.

One command, "!cclimit," even produces the spending limit on a particular card number, according to the report. Where that information comes from is unclear; the report's authors believe some of the bots are interfacing in real time with credit card company databases. "That's what we're particularly interested in," says McCarty. "They seem to have an automated system for doing that."

The Research Alliance's monitoring also produced logs of corrupt merchants offering to sell large quantities of card numbers for a percentage of the take, though the report concluded that bulk transfers were handled in private chats, or outside of IRC.

The channels named in the report have since been shutdown by DALnet, says McCarty. "We've turned a substantial amount of data over to the FBI," he adds.