Digital experience platform Optimizely has patched a vulnerability that hackers exploited to inject malicious code into customer projects, causing the websites of many major organizations to serve malware to visitors.
The company confirmed the incident in a status page update posted early on November 17 and in a post on X, where it said some Web Experimentation customers were affected.
“Earlier today, Optimizely identified a security incident in which an unauthorized party exploited a vulnerability in Experimentation and used it to add malicious JavaScript to some customer projects,” the company said on its status page. “We’ve fixed the vulnerability and removed the injected code from affected projects.”
Optimizely added that it has taken steps to secure affected user accounts and introduced additional protections, but it has not said how many customers were impacted.
Report URI, a security firm that investigated the attack, said it examined 1,611 Optimizely scripts and identified 188 infected ones, belonging to 178 organizations. Names on the list include the BBC, CNN, Zoom, GitLab, Salesforce, Equifax, HSBC and IKEA.
According to Report URI, the malicious code was placed in an optional field meant for a customer’s own JavaScript, which executes on every page. The tampered files came from Optimizely’s content delivery network, so the websites loading them treated them as trusted.
The first Optimizely injection in Report URI’s timeline is dated November 16, 04:21 UTC. By 21:52 UTC, only one of the infected scripts had yet to be cleaned.
The malware could swap cryptocurrency wallet addresses for the attacker’s own and skim payment card details. It could also collect cookies and screenshots of the page. By Monday evening, roughly $7,000 in cryptocurrency had reached the attacker, Report URI said.
iPhone and iPad users were served the Coruna and DarkSword iOS exploit kits, according to Report URI. Hudson Rock confirmed links to the exploit-kit infrastructure, but noted that successful exploitation has not been demonstrated.
“The supported conclusion is a connection to exploit-kit delivery infrastructure. The current evidence does not establish a zero-day, successful device exploitation, or collection of specific device data by unseen downstream stages,” Hudson Rock said.
Hudson Rock’s report, which includes findings from a researcher at Ohey, shows that Optimizely was not the only delivery channel. Scripts from Kissmetrics, StorifyMe and Minute-ly were seen serving a small loader that led to the same iOS-targeting infrastructure.
The earliest of these observations dates to November 13. It’s unclear how the code got into these providers’ scripts.
Hudson Rock also warned that blocking the attacker’s domain may not be enough. Some of the poisoned Optimizely scripts had a full version of the payload embedded in them, which could run if the remote file failed to load.
“The important lesson here is that trusted JavaScript doesn’t always remain trustworthy. A website can be compromised through a third-party dependency without its own code or infrastructure ever being breached,” Report URI’s Scott Helme said.
Optimizely disclosed a separate incident in February, when a threat actor accessed its systems through a voice-phishing attack.
Related: Barracuda Hacked Amid Surge in Artex-Powered Cyberattacks
Related: Microsoft Rules Out Cyberattack in Outage, Warns of Phishing Campaign
