SignDrop: The Windows CLFS Zero-Day That Turns a Security Feature Into a SYSTEM Shell — and Bypasses Defender on the Way

SignDrop: The Windows CLFS Zero-Day That Turns a Security Feature Into a SYSTEM Shell — and Bypasses Defender on the Way

SignDrop: The Windows CLFS Zero-Day That Turns a Security Feature Into a SYSTEM Shell — and Bypasses Defender on the Way

Windows privilege escalation has a new entry on its highlight reel. On Tuesday, November 10, a security researcher publishing as MSNightmare — identified elsewhere as Abdelhamid Naceri, also known as Nightmare Eclipse — dropped SignDrop, a proof-of-concept exploit with a twist: it doesn't attack the Common Log File System (CLFS) driver the way ten KEV-catalogued CLFS flaws before it did. It attacks the authentication feature Microsoft added in October 2025 specifically to stop those exploits, and it drags Windows Defender into the chain as the privileged process that does the heavy lifting.

The mechanics are precise. CLFS is Windows' kernel-mode logging subsystem, and its driver, clfs.sys, has been a favorite privilege-escalation target for years — 10 CLFS flaws already sit in CISA's Known Exploited Vulnerabilities catalog. Microsoft's October 2025 answer was logfile authentication: integrity codes checked when privileged processes open prepared logs. SignDrop abuses the moment the driver creates the authentication companion file (the .cnpf file). A standard user prepares a CLFS log and container pointing at paths they control; when a privileged process later opens that log, the driver creates the companion file in the privileged process's security context. The exploit uses object-manager symbolic links to redirect that file creation to an attacker-chosen location — and because the .cnpf file inherits the original log's permissions, the result is a SYSTEM-owned file the standard user can write to, which the researcher chains into a full SYSTEM shell.

The published chain coerces the Defender antimalware service into the privileged role through its signature-update RPC interface, and the repository claims the technique generalizes to any antimalware engine running with elevated privileges. Tested on Windows 11 26H2 and Windows Server 2025, SignDrop is claimed to reach roughly 80% of the world's Windows installations — every installation older than about 90 days, since newer systems sit in the CLFS authentication "learning" mode and typically transition to the exploitable enforced mode automatically. Korean firm 78researchlab independently compiled and tested the exploit and confirmed it works; Cyderes did the same. Microsoft has not confirmed the flaw, assigned a CVE, or shipped a fix — though the company says Defender currently detects and blocks exploitation, and no in-the-wild attacks have been observed. One consolation: the published exploit makes every UAC prompt spawn both a user and a SYSTEM shell, making an attack very visible to anyone at the keyboard.

What Should You Do?

  1. Don't wait for a CVE. Microsoft hasn't assigned one yet, so treat the Microsoft Defender signature update that blocks the chain as your mitigation — confirm your endpoints are pulling current signatures, not frozen at an old definition set.
  2. Check your CLFS authentication registry mode. Systems in enforced or learning mode (registry values 0 or 1) are inside the claimed attack surface; know which mode your fleet is in before a patch lands.
  3. Watch the console, not just the logs. The PoC blocks all other UAC actions while it runs — a user reporting a "frozen" prompt is a live indicator, so train helpdesk to escalate that specific symptom.
  4. Reduce standing local-admin friction anyway. Every Windows LPE, including this one, needs code execution as a standard user first — application allow-listing and restricted privilege keep the whole class of CLFS escalations from mattering.

The WAF Angle

SignDrop is an endpoint story, but it teaches a lesson web defenders keep re-learning: a security control added on top of a vulnerable surface creates a new surface of its own. Microsoft hardened CLFS with logfile authentication; the attacker simply found the code path where the new control creates a file with inherited permissions. This is the exact pattern of WAF bypasses built on the very features meant to protect — a WAF that special-cases a trusted parameter, an RMM that whitelists its own update path, an auth flow whose "secure" recovery branch skips the rate limiter. The lesson is to audit the control's own mechanics: every exception, inherited-permission path, and privileged-context operation your defenses introduce. CLFS itself has been an attacker staircase for years — as we covered when PHP2Shell chains and other Windows kernel escalations made patch hygiene the only reliable floor.

Sources