Optimizely supply-chain attack injects malware into major brand sites
Malicious JavaScript served via Optimizely Web compromised 178 organizations including Zoom and BBC, stealing crypto and payment data on November 16, 2026.
On November 16, 2026, a sophisticated supply-chain attack injected malicious JavaScript into the web snippets of 178 major organizations through the Optimizely Web experimentation platform. High-profile victims included Zoom, the BBC, CNN, Salesforce, GitLab, IKEA, HSBC, Equifax, Air France-KLM, Ledger, and Peacock. The incident highlights how trusted third-party dependencies can become vectors for widespread compromise without any breach of the victim's own infrastructure.
What happened
Optimizely Web functions by having customers load a JavaScript snippet from Optimizely’s content delivery network. This snippet contains not only the platform’s core code but also a field called Project JavaScript, which allows customers to insert custom code that runs on every page view. In this incident, an attacker gained write access to multiple customer projects and inserted malicious code directly into this Project JavaScript field. Because the resulting file was still served from cdn.optimizely.com with valid certificates, browsers and security tools treated it as trusted content.
The injected code acted as a loader, fetching further payloads from attacker-controlled domains such as 1688360.cc. Researchers observed three variations of the injection: a plain script tag pointing to the malicious host, a base64-encoded version to evade simple text searches, and a full inline copy of the payload that executed via eval if external connections were blocked. The attack affected even those customers who hosted their snippets on their own domains, as the content originated from the compromised Optimizely projects. Optimizely’s own corporate sites were also infected, confirming the issue lay within the platform’s project management rather than just individual customer accounts.
The timeline reveals a rapid escalation. The first injection was detected in the BBC’s snippet at 04:21 UTC on November 16. By 08:28 UTC, the primary malicious script track.js was active on bbc.com. Throughout the day, the attacker rebuilt the payload 26 times, adding features like card skimming and iOS exploit kits while never removing previous capabilities. A mass cleanup effort began around 20:40 UTC, and by 21:52 UTC, only one of the 188 scanned snippets remained infected. The method by which the attacker gained cross-customer write access remains unconfirmed, though possibilities include compromised credentials or a platform-level vulnerability.
Key details
- Scope: 178 organizations were compromised, including major brands like Zoom, HSBC, and the BBC.
- Vector: Malicious code was injected into the Project JavaScript field of Optimizely Web projects.
- Payload: The script performed crypto address swapping, credit card skimming, cookie theft, and screenshot capture.
- iOS Exploits: Visitors on iPhones and iPads running older iOS versions were targeted with Coruna and DarkSword exploit kits.
- Domains: Key attacker infrastructure included
1688360.cc,5vobauua3ojpklx.icu, andwebshell.lol. - Mitigation: Most infections were cleaned by 21:52 UTC on November 16, but data exposure likely occurred during the active window.
Background
This incident is a classic example of a supply-chain attack, where attackers compromise a trusted vendor to reach its customers. In web development, organizations frequently rely on third-party scripts for analytics, A/B testing, and marketing. These scripts run with the same privileges as the site’s own code, meaning a compromise in one provider can affect all its clients simultaneously. Unlike a direct breach of a company’s server, this type of attack bypasses traditional perimeter defenses because the malicious code comes from a domain the browser already trusts.
Content Security Policy (CSP) is a browser security standard designed to mitigate such risks by specifying which sources of content are allowed to load. A strict CSP can block scripts from unauthorized domains or prevent inline code execution. However, many organizations configure CSPs loosely, often allowing unsafe-eval or using strict-dynamic, which can inadvertently trust any script loaded by a permitted parent script. In this case, a restrictive CSP could have blocked the malicious connections or script execution, but most affected sites lacked such protections.
Why it matters
For teams that manage their own software infrastructure, this event underscores the hidden risk of third-party dependencies. Even if your internal code is secure and your servers are hardened, loading a script from a compromised vendor can expose your users to malware. The attack did not require exploiting a vulnerability in the victim’s application; it exploited the trust relationship between the customer and Optimizely. This means that security monitoring must extend beyond your own codebase to include the behavior of external scripts running in your users' browsers.
The impact on user trust and financial liability is significant. The malware actively swapped cryptocurrency wallet addresses, leading to verified theft of funds. It also skimmed credit card details entered into non-hosted forms and stole session cookies. For IT and security leads, this raises difficult questions about incident response and customer communication. Determining which users were affected requires detailed logging of script loads and user interactions, which many organizations may not have in place. Furthermore, the use of iOS exploit kits suggests that the attackers were capable of delivering more severe device-level compromises, increasing the potential severity of the incident.
What you can do
- Audit third-party scripts: Review all external JavaScript loaders, particularly those from A/B testing or marketing platforms, for unexpected changes or unknown code.
- Implement strict CSP: Deploy a Content Security Policy that restricts
script-srcto known hosts and avoidsunsafe-evalorstrict-dynamicunless absolutely necessary. - Rotate credentials: If you use Optimizely or similar platforms, rotate all API tokens and review user access permissions immediately to rule out compromised accounts.
- Monitor for IOCs: Block known malicious domains such as
1688360.ccand5vobauua3ojpklx.icuat your firewall or DNS level. - Check for data exposure: Treat any credit card entries or cryptocurrency transactions made on affected sites on November 16 as potentially compromised and monitor for fraud.
- Update devices: Users who visited affected sites should update their iOS devices to the latest version to patch any potential exploit kit vulnerabilities.



