On the very day Microsoft delivered a record-setting batch of security fixes, a security researcher publicly released exploit code for a previously unknown Windows vulnerability. The timing has intensified scrutiny of the software giant's vulnerability disclosure practices and left administrators scrambling once again.
Microsoft's latest Patch Tuesday addressed a staggering 570 security flaws, the highest number the company has ever tackled in a single update cycle. Yet within hours, the spotlight shifted from that milestone to a fresh zero-day exploit published by a researcher operating under the pseudonym NightmareEclypse.
Nine Exploits and Counting
NightmareEclypse has now released nine such exploits, each accompanied by complaints about how Microsoft handles their bug reports. The latest, dubbed HiveLegacy, was disclosed on Tuesday. According to the researcher, the proof-of-concept code included in the publication was deliberately stripped down to reduce the likelihood that attackers could weaponize it immediately.
Multiple independent security researchers have confirmed that the exploit functions as described. The confirmation has put additional pressure on Microsoft to develop and distribute a patch, even as the company processes the enormous volume of fixes from its record Patch Tuesday.
How HiveLegacy Works
HiveLegacy is classified as an elevation-of-privilege exploit. It takes aim at a vulnerability within the Windows User Profile Service, a core operating system component. By exploiting this flaw, an account with limited system permissions can compromise a higher-privilege administrator account.
The attack works by modifying the targeted admin account's classes registry hive. This registry resource plays a key role in determining which application launches when a user clicks on specific file types within Windows Explorer. By tampering with it, an attacker with low-level access could effectively hijack administrative functions.
While the current proof-of-concept demonstrates the exploit working at the user level, researchers note that with additional effort, the same underlying technique could potentially be adapted to compromise automated processes running on the system as well.
