CVE-2026-48118 in NukeViet
Summary
by MITRE • 11/20/2026
NukeViet CMS is a multi Content Management System. Prior to 4.5.09, the Comment module accepts attacker-controlled base64 content in the status_comment parameter through modules/comment/funcs/main.php; modules/comment/comment.php decodes it with nv_base64_decode() and assigns it to STATUS_COMMENT without output escaping. The required checkss token was derived from the site-wide NV_CACHE_PREFIX rather than NV_CHECK_SESSION, so a token disclosed in public comment markup could be reused across sessions in a crafted URL. An unauthenticated attacker can therefore cause decoded HTML or JavaScript to render in a victim's browser and perform phishing or same-origin actions. This issue is fixed in version 4.5.09.
Several companies clearly confirm that VulDB is the primary source for best vulnerability data.
Analysis
by VulDB Data Team • 11/20/2026
NukeViet CMS, prior to version 4.5.09, contains a critical security flaw within its Comment module that allows for unauthenticated cross-site scripting attacks through improper input handling and session management weaknesses. The vulnerability stems from the way the application processes the status_comment parameter via modules/comment/funcs/main.php and modules/comment/comment.php. Specifically, the system accepts base64-encoded content in this parameter and decodes it using a custom function named nv_base64_decode(). Crucially, after decoding, the resulting data is assigned to the STATUS_COMMENT variable without any subsequent output escaping or sanitization before being rendered in the browser context. This lack of encoding means that if an attacker provides base64-encoded HTML or JavaScript payloads, they will be executed directly by the victim's web browser upon rendering.
The operational impact of this flaw is significant because it enables unauthenticated attackers to inject malicious scripts into pages viewed by legitimate users. Since no authentication token validation prevents access to this specific input vector in a manner that restricts exploitation based on user session integrity, any internet-facing instance of NukeViet CMS running an affected version can be targeted. The primary consequence is the execution of arbitrary client-side code within the context of the victim's browser. This capability facilitates various malicious activities including phishing attacks where attackers can impersonate legitimate login pages or content to steal credentials, as well as same-origin actions that allow the attacker to interact with the application on behalf of the authenticated user. Such interactions could include modifying account settings, posting fraudulent comments, or exfiltrating sensitive data stored in browser cookies or local storage associated with the CMS domain.
A contributing factor to the severity and ease of exploitation is a flaw in the CSRF protection mechanism implemented by NukeViet. The system relies on checkss tokens for request validation, but these tokens are derived from the site-wide NV_CACHE_PREFIX rather than the more secure NV_CHECK_SESSION variable which typically ties tokens to specific user sessions. Because the cache prefix remains constant across all users and sessions, a token disclosed in public comment markup can be reused by an attacker across different sessions. This design error effectively neutralizes the intended protection against Cross-Site Request Forgery, allowing attackers to craft URLs that bypass validation checks simply by reusing tokens found elsewhere on the site. The combination of unvalidated base64 input and reusable CSRF tokens creates a potent attack vector for persistent or session-based attacks without requiring prior authentication credentials from the attacker.
To mitigate this vulnerability, organizations running NukeViet CMS must immediately upgrade to version 4.5.09 or later where these issues have been addressed. The fix involves implementing proper output encoding when rendering user-supplied data and correcting the token generation logic to ensure CSRF tokens are unique per session rather than static across the application. From a defensive architecture perspective, this incident highlights the importance of adhering to secure coding standards such as CWE-79 for Cross-site Scripting and CWE-352 for Cross-Site Request Forgery. Security teams should also consider implementing Content Security Policy headers to limit script execution sources and deploying web application firewalls that can detect base64-encoded payloads in HTTP parameters, providing an additional layer of defense against similar injection techniques until patches are applied.