Optimizely Supply Chain Attack Injects Card Skimmers and iOS Exploit Kits into Scripts Serving CNN, BBC and Zoom

An attacker exploited a vulnerability in Optimizely Web Experimentation to inject malicious JavaScript into customer project snippets served from cdn.optimizely.com. The code reached visitors of at least 178 organizations including CNN, BBC, Zoom, Salesforce, GitLab, IKEA, HSBC and Equifax. The payload performed crypto wallet address replacement, captured card details from forms, and on iOS devices attempted to load the Coruna and DarkSword exploit kits. Optimizely stated it has patched the flaw and removed the unauthorized code, though some projects were re-infected after initial cleanup. Report URI, Cynet and StepSecurity published detailed analyses showing the loader used multiple obfuscation techniques and delivered track.js, card.php and boot.js components. Crypto wallets linked to the campaign received roughly $7,000 within nine hours. The incident highlights risks of third-party experimentation platforms that serve executable code to high-traffic websites.

The Hacker NewsTranslated from English

Read at source

An attacker exploited a flaw in Optimizely Web Experimentation to plant malicious code in scripts loaded by customer websites. Visitors to sites operated by at least 178 organizations, including CNN, the BBC and Zoom, received the injected code on 17 November 2026, according to Report URI.

The injected JavaScript was designed to replace cryptocurrency wallet addresses with attacker-controlled addresses and to capture credit-card details submitted through site forms. On iPhones and iPads the code additionally attempted to load a second-stage payload identified as the Coruna and DarkSword iOS exploit kits, although no successful device compromises were observed.

Optimizely confirmed the attack in an incident notice posted at 01:02 UTC, stating that the adversary had exploited a vulnerability in the Experimentation product to add malicious Project JavaScript to customer projects. The vendor removed the unauthorized code, secured affected accounts and introduced additional protections against future unauthorized modifications.

Report URI examined 1,611 Optimizely snippets and found malicious loaders in 188 of them belonging to 178 distinct organizations. The list includes Salesforce, GitLab, IKEA, HSBC, Equifax, Air France-KLM, Ledger and Peacock. The vendor’s own main site and login page were also affected.

The loader appeared in three variants: direct inclusion of 1688360[.]cc/track.js, a base64-obfuscated version, and a fully embedded copy of the payload. The script performed address swapping only when a valid wallet address appeared near payment-related terms and exfiltrated cookies and page screenshots after a successful swap.

A card-skimming component added later captured card number, expiry date and CVV from form submissions and transmitted the data to attacker-controlled endpoints. The skimmer could not access values entered inside cross-origin payment iframes such as those provided by Stripe or Adyen.

On iOS devices the code loaded boot.js from two .icu domains that fingerprinted the operating-system version and attempted exploitation via Coruna (iOS 13.0–17.2.1) or DarkSword (iOS 18.4–18.7). Researchers from Cynet retrieved the second stage using an emulated browser and described the activity as conditional delivery rather than confirmed compromise of physical devices.

StepSecurity and Report URI recommend that affected organizations review all Project JavaScript, rotate credentials, enforce strict Content-Security-Policy rules, and treat any card details or non-HttpOnly cookies collected on 17 November as potentially exposed. Visitors are advised to update iOS devices to the latest patched versions and to verify any cryptocurrency addresses used that day.

Meridian digestThe week's most consequential security stories, distilled into one briefing.See the digest

More in Supply Chain & OSS

StepSecurity Raises $20 Million Series A to Strengthen Software Supply Chain Defenses
Microsoft Agent Framework PyPI Package Compromised with Credential-Stealing Malware
Malicious Microsoft Agent Framework Releases on PyPI Steal Cloud and CI/CD Credentials