Opens in a new tab

Optimizely Attack Exposed Visitors at Major Sites

Optimizely snippet hijacked: 178 sites serve skimmer and iOS exploits

The Optimizely attack exposed visitors to major websites to malicious JavaScript delivered through a trusted experimentation service. Active on 16 November 2026, the campaign supported cryptocurrency theft, payment card skimming, data collection and an iOS-specific second stage.

Optimizely confirmed at 01:02 UTC on 17 November that an unauthorised party had exploited a vulnerability in its Experimentation product. The company said malicious code was added to some customer projects, but that the vulnerability had since been fixed, the injected code removed and additional protections applied.

How the Optimizely attack reached major websites

The affected product was Optimizely Web Experimentation, which allows organisations to run tests and personalised website experiences. Customers typically place an Optimizely JavaScript snippet on their pages, with the browser retrieving project code from cdn.optimizely.com.

Within each project is a feature called Project JavaScript. This code runs as part of the standard snippet and can affect experiments across the project. Attackers gained the ability to modify this area in some customer projects, although Optimizely has not publicly disclosed the vulnerability’s root cause or how the attackers initially obtained access.

The malicious addition was a small loader that created a script element and directed the visitor’s browser to https://1688360.cc/track.js. The 1688360.cc domain was registered on 16 November, aligning with the first observed malicious activity and public reports.

Because the altered snippet still came from Optimizely’s genuine content delivery infrastructure, it appeared to browsers as an expected third-party resource. The customer’s own website did not need to be directly compromised. When someone opened an affected page, the trusted snippet ran in the page context and fetched the attacker’s payload.

Researchers identified several loader variations during the Optimizely attack. Some plainly referenced 1688360.cc, while others encoded the address with base64 and decoded it using the JavaScript atob function. Reports also described inline payload code that could continue operating if the external domain was blocked.

Self-hosting or proxying an Optimizely snippet did not necessarily prevent exposure. If the copied content came from a compromised project, the malicious Project JavaScript could still be distributed through the organisation’s own domain.

Which organisations and products were affected

Sites named in scans or reporting included the BBC, CNN, Zoom, Salesforce and GitLab. Other reported names included IKEA, HSBC, Equifax, Air France-KLM and Ledger. The presence of a site in scanning data indicates that an Optimizely-served page or snippet was observed loading the malicious code, but it does not establish how many visitors were exposed or whether data was successfully stolen from each organisation.

The total also remains unsettled. Several reports cite 178 affected organisations, while StepSecurity observed 186 sites. Optimizely has not published an authoritative customer list, so both the final number and the duration of exposure for individual sites remain unconfirmed.

No affected Optimizely version numbers have been released. The relevant scope is customer projects in Web Experimentation where malicious content was inserted into Project JavaScript and distributed within an Optimizely snippet. Any website platform or content management system embedding one of those project snippets could therefore have delivered the payload.

What the malicious JavaScript did

Cryptocurrency address replacement

The track.js payload searched pages, input fields, clipboard activity and some outbound requests for cryptocurrency wallet addresses. When it found a supported address, it could replace it with one controlled by the attacker, causing a user to send funds to the wrong destination without noticing the change.

Researchers observed support for several wallet formats, including Bitcoin and Ethereum. Independent estimates placed the funds received by identified attacker wallets at approximately USD 6,600 to USD 7,000 during the observed period. These figures are provisional and may not represent the campaign’s complete financial impact.

Payment card skimming

The Optimizely attack also introduced a card skimmer that monitored forms for card numbers, expiry dates and security codes. Captured details were sent towards 1688360.cc/card.php.

The skimmer was most relevant where card details were entered into forms controlled directly by the affected website. Hosted payment fields and cross-origin payment iframes, such as those operated by dedicated payment providers, can isolate card entry from scripts running on the surrounding page and therefore limit this specific collection method.

Cookies, URLs and screenshots

The malware collected further context, including page URLs, cookies readable by JavaScript and screenshots of displayed pages. This could expose session information, customer details or other content visible during a visit, depending on the affected page and the protections applied to its cookies.

A captured track.js build had a SHA-256 hash beginning f7b84b9a and was observed returning successfully to multiple sites between 17:57 and 21:23 UTC on 16 November. As the attackers revised their code, a single hash should not be treated as a complete detection method.

iPhones and iPads received a separate stage

On Apple mobile devices, the payload contacted additional infrastructure using .icu domains and fetched an iOS-specific second stage. Researchers confirmed that routing varied according to device context, indicating that the attackers were attempting more than web form theft on iPhones and iPads.

Some reports associated this branch with the Coruna and DarkSword exploit kits. However, that attribution has not been confirmed consistently by vendor-grade evidence. The reliable conclusion is that an additional iOS-targeting stage existed; the specific exploit framework should remain classified as unconfirmed.

Optimizely attack timeline and current status

  • 16 November 2026: The 1688360.cc domain was registered, and public scans captured Optimizely-served pages loading track.js from that domain.
  • 16 November 2026: The first public investigation described the injected loader, named the campaign OPTIMJACK and documented its theft capabilities.
  • 16 to 17 November 2026: Independent researchers observed payload changes, affected sites and funds entering attacker-controlled cryptocurrency wallets.
  • 17 November 2026 at 01:02 UTC: Optimizely confirmed the security incident, said the exploited vulnerability was fixed and reported that injected code had been removed.

This was confirmed exploitation in the wild, not a theoretical proof of concept. The genuine snippets were modified and executed automatically when visitors loaded affected pages. No threat actor has been publicly identified.

Optimizely’s remediation should stop the known delivery route, but it cannot determine whether information was taken from browsers before removal. Affected organisations must use their own content security policy reports, DNS data, proxy records, browser telemetry and payment monitoring to establish local impact.

Checks organisations should make now

Organisations using Optimizely Web Experimentation should retrieve and inspect the exact snippet loaded by their live pages, rather than relying only on stored source code. Priority actions directly connected to the Optimizely attack include:

  • Search snippets and historical telemetry for 1688360.cc, track.js, card.php and base64 or atob code resolving to the same infrastructure.
  • Block the confirmed domain and identified .icu second-stage hosts through content security policy, DNS and web proxy controls.
  • Review outbound requests, readable cookie exposure and content security policy reports from 16 November.
  • For first-party payment forms, investigate suspicious submissions and compare findings with payment provider or card-brand alerts.
  • Contact Optimizely support and monitor its incident updates if project compromise or unexplained Project JavaScript changes are suspected.

The incident demonstrates why third-party experimentation code must be treated as active production code. For organisations potentially affected, the immediate priority is establishing whether the malicious snippet reached visitors and whether browser data left their websites before Optimizely completed remediation.

Originally reported by cybersecuritynews.com.

Share this bulletin

About the Author

Rob McBride Headshot - CyPro Partner and leading cyber security expert

Rob McBride

Partner

  • CISSP
  • ACA Chartered Accountant
  • MPhil
  • BSc
  • SOC 2
  • ISO 27001

Rob McBride

Rob is a Founding Partner at CyPro and a highly experienced CISO. Beginning his career with a successful tenure at Deloitte, Rob has since amassed a wealth of experience, notably serving as a cyber security advisor to the UK government and spearheading cloud security transformations for several global banks.

At CyPro, Rob leads the managed service business line, working extensively across multiple sectors including telecommunications, technology, higher education, travel, and retail. He is passionate about equipping small and medium-sized businesses (SMBs) with robust cyber security strategies to fuel their growth.

View Profile
Back to Bulletins

Related CyPro Services

  • Managed Detection and Response (MDR)

    Managed Detection and Response (MDR) is an end-to-end managed service designed to help organisations detect, analyse and respond to cyber threats quickly and effectively. It...
    View Service
CyPro Cookie Consent

Hmmm cookies...

Our delicious cookies make your experience smooth and secure.

Privacy PolicyOkay, got it!

We use cookies to enhance your experience, analyse site traffic, and for marketing purposes. For more information on how we handle your personal data, please see our Privacy Policy.

Schedule a Call