DarkSword iOS exploit is now drawing serious attention after researchers confirmed that the advanced exploit chain is no longer limited to its original operators.
This development raises serious concerns because it means a highly sophisticated exploit is no longer restricted to a limited group of attackers.
A recent demonstration by security researcher @matteyeux showed successful kernel read and write access on an iPad mini 6 running iOS 18.6.2 using the DarkSword exploit. This public validation shows that the exploit remains effective in real-world conditions and increases the risk for millions of Apple devices that have not yet been patched.
Google Threat Intelligence Group reportedly first observed DarkSword in active campaigns in November 2025. The exploit kit has been mainly linked to UNC6353, a suspected Russian espionage group that previously used the Coruna iOS exploit kit. Reported targets have included victims in Ukraine, Saudi Arabia, Turkey, and Malaysia, showing that the threat has already been used in focused international operations.
Technical Structure and Post-Compromise Activity
DarkSword is not just a single exploit but a complete exploit kit and infostealer written in JavaScript. The attack typically begins when a victim visits a compromised website containing a malicious iframe, a method commonly associated with watering hole attacks.
Once the target opens the page, the exploit escapes Safari’s WebContent sandbox. It then bypasses important Apple protections, including Trusted Path Read-Only and Pointer Authentication Codes, by abusing sensitive internal dyld structures in writable stack memory. The chain then moves through the GPU process by exploiting an out-of-bounds write flaw in the ANGLE graphics engine before targeting the XNU kernel through a Copy-On-Write vulnerability in the AppleM2ScalerCSCDriver driver.
This gives attackers arbitrary memory read and write access, allowing them to modify sandbox restrictions and reach protected parts of the file system. Researchers also found that DarkSword operates fully in memory and quickly loads final-stage malware after compromise. Three malware families linked to the activity have been identified as GHOSTBLADE, GHOSTKNIFE, and GHOSTSABER. These payloads are designed to steal sensitive data, including secure messages, saved credentials, and cryptocurrency wallet information.
Security Response and Protection Measures
The public validation of DarkSword by independent researchers significantly increases the overall threat level. Once a working exploit chain becomes accessible beyond its original operators, the chances of wider abuse rise sharply.
The command-and-control infrastructure used in these operations adds to the concern. Instead of using obvious malicious domains, attackers relied on subdomains created on compromised legitimate websites, helping their traffic blend in and making detection harder.
To reduce risk, Apple users and enterprise security teams should ensure that all devices are updated immediately to iOS 26.1 or later, as these versions include fixes for the kernel vulnerabilities involved in the exploit chain. For high-risk users such as journalists, executives, and government personnel, enabling Apple’s Lockdown Mode can provide an additional layer of defense against advanced web-based attacks.