Bandizip EUVD-2026-105087

Requires that (1) the victim uses Bandizip before 7.47 to extract an archive containing a .ps1 or .psm1 file, (2) the Windows host's PowerShell execution policy is set to RemoteSigned (so that the normally-blocking zone check is what the bypass defeats), and (3) the user subsequently and manually invokes the extracted script. Exploitation is limited by its local attack vector (the archive must reach the victim's disk) and by required user interaction - the victim must both extract and run the script - so it is not remotely or automatically triggerable; it also provides no benefit on hosts set to Unrestricted (nothing to bypass) or AllSigned/Restricted (bypass still blocked).

This is a genuine-but-modest-priority issue, not a top-tier emergency. The CVSS 4.0 vector (AV:L/AC:L/AT:N/PR:N/UI:P, score 4.8 MEDIUM) is internally consistent with the description: exploitation is local (the malicious archive must already be on disk and extracted by Bandizip) and requires passive user interaction (UI:P) because the victim must subsequently invoke the extracted PowerShell script. Impact is correctly scoped to integrity only (VI:L; VC:N/VA:N), reflecting that this is a protection-mechanism bypass (CWE-693) that weakens the Mark-of-the-Web defense-in-depth layer rather than directly granting code execution. No EPSS score, CISA KEV listing, or public POC was provided in the input, so there is no signal of active exploitation or automatable mass abuse - consistent with a MEDIUM-severity defense-weakening bug reported by a research vendor (VulnCheck) rather than one caught in the wild. One nuance to verify: the real-world severity depends on the host PowerShell execution policy being RemoteSigned (a common but not universal default); on systems set to Restricted or AllSigned the bypass yields less, and on systems already at Unrestricted the MotW gap is moot. Treat as a patch-in-normal-cycle item.

An attacker distributes a ZIP (or other) archive containing an unsigned malicious .ps1 or .psm1 file via download, email, or a shared link. Because Bandizip before 7.47 does not propagate Mark-of-the-Web to the extracted scripts, the victim who extracts and then runs the script sees it treated as local/trusted content and it executes under a RemoteSigned policy that would normally block an unsigned downloaded script. No public exploit code was identified in the provided data; exploitation requires the user to both extract and deliberately invoke the script.