Skip to content
Quantum Secure Labs
Cybersecurity News

Optimizely flaw injected a card skimmer and iOS exploits into websites of 178 organisations

On 16 November, an attacker exploited a flaw in Optimizely Web Experimentation and injected malicious JavaScript into websites of at least 178 organisations: card theft, crypto wallet swapping and iOS exploits. If your…

By Quantum Secure Labs4 min read
Pantalla de ordenador con código JavaScript en un editor
In this article

On 16 November, an attacker exploited a flaw in Optimizely Web Experimentation and injected malicious JavaScript into websites of at least 178 organisations: card theft, crypto wallet swapping and iOS exploits. If your website uses Optimizely, check your project code today and rotate credentials. If not, audit your third-party scripts.

Key facts

  • The first injection recorded by Report URI dates from 16 November at 04:21 UTC, in the BBC's script.
  • Report URI reviewed 1,611 Optimizely scripts and found 188 infected, belonging to 178 organisations, including BBC, CNN, Zoom, GitLab, Salesforce, IKEA, HSBC, Equifax and Air France-KLM. It estimates the real figure is higher (BleepingComputer).
  • The code was placed in the Project JavaScript field, which runs on every page and was served from Optimizely's CDN.
  • By 19:25 UTC on the 16th, the attacker's wallets had received around $7,040. It is unknown how many transfers came from victims.
  • Mass clean-up began around 20:40 UTC. By 21:52 UTC only TalkTalk's script was still infected. Smartsheet was reinfected after the first clean-up.

What did the malicious code do?

According to SecurityWeek, the vulnerability was exploited to serve malware to visitors of major websites. The payload, track.js, did three things:

  • Replaced crypto wallet addresses with the attacker's.
  • Captured card number, expiry date and CVV submitted through the site's own forms. It could not read cards entered in cross-origin payment iframes, such as Stripe or Adyen hosted fields.
  • Sent out cookies and screenshots.

On iPhone and iPad it loaded two exploit kits: Coruna (iOS 13.0–17.2.1) and DarkSword (iOS 18.4–18.7). Hudson Rock and Cynet note that no successful device exploitation has been demonstrated.

What has Optimizely said?

Optimizely told BleepingComputer that the attacker exploited a vulnerability in the Experimentation account sign-up flow. It says it blocked that path, deployed a fix, removed the injected code, terminated sessions, revoked tokens and notified affected customers.

The problem is what's missing. According to The Hacker News, its notice (01:02 UTC on 17 November) does not explain how the flaw worked, how many projects were modified or what customers should do. You'll have to ask.

Does this affect me if I don't use Optimizely?

Not directly. But Hudson Rock saw similar loaders in scripts from Kissmetrics, StorifyMe and Minute-ly, with sightings since 13 November. It is unclear how the code reached those vendors.

The other lesson is about defences. According to Report URI's analysis cited by BleepingComputer, 68 of the affected sites had no Content Security Policy (CSP), and others had one so permissive that the script ran anyway. Note that allowing cdn.optimizely.com would not have helped, because the code came from that very host.

What to do today

  1. If you use Optimizely, open the Project JavaScript of every project and the custom code of every experiment. Look for "1688360", "atob(" or any script you didn't write.
  2. Review which users, tokens and integrations can write to the account and rotate their credentials.
  3. Ask Optimizely in writing which projects were modified, when and how.
  4. If your forms collect cards, consider those entered on Monday 16th exposed. Treat session tokens stored in cookies without HttpOnly as exposed.
  5. Update company iPhones and iPads that may have visited affected sites. The Hacker News recommends checking those on iOS 18.6 or earlier.
  6. Whether or not you use Optimizely, inventory your third-party scripts (analytics, A/B testing, chat, tags) and remove them from payment pages.

What it means for a small business

Our take: the flaw wasn't in the affected websites, but the damage was. Every third-party script you load has the same permissions as your own code. If that vendor falls, your website falls.

A small business can't audit its vendors, but it can limit what a foreign script can reach. Two measures cover most of the risk in this case. First: take payments with your payment provider's hosted fields, in a cross-origin iframe. The skimmer couldn't read them. Second: deploy a strict CSP with script-src and connect-src, no wildcards. It isn't trivial, but it's days of work, not months. If you need help with it, we cover it in cybersecurity for SMEs.

There's another question the sources don't answer: if you're an EU company and processed personal data of affected customers, assess with your DPO or adviser whether you need to notify a breach under the GDPR. This assessment is ours.

FAQ

Is the Optimizely flaw fixed?

Optimizely says it fixed the vulnerability and removed the injected code. Even so, Smartsheet was reinfected after the first clean-up, so check your projects yourself.

Am I safe if I take payments with Stripe or Adyen?

If you use their hosted fields in a cross-origin iframe, the payload could not read the cards. If your form is your own, no.

Is having a CSP enough?

Not if it's permissive. In this attack, allowing cdn.optimizely.com offered no protection, because the malicious code came from that host.

Sources

Worried about a threat like this one?

We review your exposure and show you how to protect yourself.

See our cybersecurity service

Cybersecurity News

PostgreSQL fixes 52 security flaws, one of which lets attackers run code on the server

PostgreSQL released updates on 12 November fixing 52 security flaws. The most serious, CVE-2026-6460 (CVSS 8.8), lets a user with permission to create objects run code on the server. It affects every version before…

Cybersecurity News

Citrix patches critical memory flaws in NetScaler ADC and Gateway: time to update

Citrix has released patches for around 28 vulnerabilities in NetScaler ADC and NetScaler Gateway, nine of them critical memory overflow flaws. It affects companies using NetScaler as a VPN, remote access gateway or load…

Cybersecurity News

Check Point ships emergency hotfixes as attackers exploit gateways with PHP2Shell and a root flaw

Check Point has released emergency hotfixes because attackers have been exploiting unpatched Security Gateways, from R80 to R82.20, since late October to gain root. If your company uses Check Point firewalls or VPNs…

Want to apply these ideas to your business?

Tell us about your case and we'll show you how to apply these strategies to your project.