CVE-2025-9491 is a remote code execution vulnerability in Microsoft Windows involving the handling of .LNK shortcut files. The flaw is a user-interface misrepresentation issue in which crafted shortcut metadata can cause hazardous content, particularly malicious command-line content associated with the shortcut target, to be hidden or truncated when viewed through the Windows-provided Properties interface. Reported exploitation commonly relied on the fact that Windows Explorer displayed only the first 260 characters of the shortcut Target field while the underlying command line could be far longer, allowing attackers to pad the visible portion and conceal malicious arguments from user inspection. When the victim opens the crafted shortcut, Windows executes the full underlying command, enabling arbitrary code execution in the security context of the current user. The issue has been associated with active in-the-wild exploitation in phishing and espionage campaigns and was also tracked as ZDI-CAN-25373 / ZDI-25-148.
Mallory correlates every CVE against your assets, your vendors, and active adversary campaigns. Know which vulnerabilities matter for you, not just which ones are loud.
What it means. What to do now. Patch path, mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
If you can’t patch tonight, do this now.
Patch, then assume compromise.
1 valid exploit after Mallory filtered fakes, detection scripts, and README-only repos.
This repository is a proof-of-concept (PoC) tool for CVE-2025-9491, a Windows UI misrepresentation vulnerability affecting .lnk (shortcut) files. The repository contains three files: a detailed README.md, the main exploit script (main.py), and a requirements.txt specifying the pylnk3 library. The main.py script is a Python tool that allows users to create new .lnk files with obfuscated (whitespace-padded) arguments, obfuscate existing .lnk files, or parse and display the contents of .lnk files. The exploit leverages the fixed-size 'Target' field in Windows shortcut properties, using whitespace padding to hide malicious command-line arguments from users inspecting the shortcut. The tool supports various padding patterns and custom character sets for obfuscation. The primary attack vector is local: the attacker must convince a user to open or inspect a malicious .lnk file, which can then execute hidden commands. The tool does not provide a remote shell or network-based payload, but enables stealthy local execution of arbitrary commands via shortcut files. The repository is well-documented and structured, with the main exploit logic contained in main.py.
Products and vendors Mallory has correlated with this vulnerability. Open in Mallory to drill down to specific CPE configurations and version ranges.
Vendor-confirmed product mapping. Mallory continuously reconciles this list against your asset inventory.
93 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
A Microsoft Windows vulnerability listed among vulnerabilities actively exploited or operationally weaponized in July 2026.
A vulnerability abused in an Armored Likho campaign to execute obfuscated PowerShell and deploy BusySnake Stealer.
A now-patched remote code execution vulnerability in the way Windows handles shortcut (LNK) files, abused in this campaign to execute an obfuscated PowerShell command and launch malware.
Windows Shortcut (LNK) UI misinterpretation vulnerability leading to remote code execution; exploited by multiple threat actors since 2017; patched in Nov 2025 updates.
Query your assets running an affected version, and investigate the blast radius.
Every observed campaign linking this CVE to a named adversary.
Malware families riding this exploit, with evidence and IOCs.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Cross-references every affected SKU, including bundled OEM variants.
Community discussion across Reddit, Mastodon, and other social sources.