Description

The TranslatePress – Translate Multilingual sites with AI Translation plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'comment' parameter in all versions up to, and including, 3.3.8 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Successful exploitation requires that at least one prior comment from the attacker's name and email pair has been approved under WordPress's default comment_previously_approved moderation setting, and that a secondary language is configured in TranslatePress.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "7.2",
  "pubDate": "2026-11-07T05:18:06.850Z",
  "pubdate": "2026-11-07T05:18:06.850Z",
  "executiveSummary": "TranslatePress – Translate Multilingual sites with AI Translation for WordPress is affected by a Stored Cross-Site Scripting (XSS) vulnerability in versions up to and including 3.3.8.\nThe vulnerability stems from inadequate sanitization and output escaping of the 'comment' parameter during the translation process.\nThis flaw allows unauthenticated attackers to inject malicious JavaScript into WordPress comments, which executes within the context of the victim's browser session when they view the translated page.\nThe impact includes potential session hijacking, unauthorized actions performed on behalf of authenticated users, or unauthorized content modification.\nExploitation requires specific conditions: the target WordPress site must have at least one secondary language configured via TranslatePress, and the attacker must satisfy the 'comment_previously_approved' moderation requirement, meaning they must have had a prior comment approved by the site administrator.\nGiven that this vulnerability permits arbitrary script execution, it poses a significant risk to site integrity and user security, necessitating immediate attention to patching or defensive configuration.",
  "technicalDetails": "The vulnerability originates in the TranslatePress plugin's handling of comment data during the rendering of multilingual content. When TranslatePress processes a page for translation, it fails to properly sanitize or escape input derived from the 'comment' parameter before rendering it in the browser.\nThe primary root cause is the lack of strict input validation and contextual output encoding on the server-side, allowing malicious payloads—typically contained within <script> tags or via event handler attributes—to be stored persistently in the database.\nThe attack flow begins when an attacker submits a comment containing a malicious script payload. Under default WordPress configurations, this comment is held for moderation. The attacker must possess a history of approved comments under the same name and email pair to bypass the 'comment_previously_approved' moderation check, which facilitates the automatic publication of the malicious payload on the site.\nOnce the payload is successfully stored, the vulnerability is triggered when a user accesses a page containing the injected comment, provided a secondary language is configured in TranslatePress. The plugin processes the comment to provide the translated version and, in doing so, improperly reflects the unsanitized script into the DOM.\nBecause the payload is stored, any visitor or administrator viewing the affected page will execute the injected script within the security context of their session. This enables the attacker to potentially exfiltrate sensitive cookies, capture CSRF tokens, or force the victim to perform unauthorized state-changing operations.\nThe exploitation is feasible by unauthenticated remote attackers who have satisfied the pre-requisite comment moderation criteria. There are no specific privilege requirements for the attacker beyond the initial post-moderation capability. The vulnerability affects all versions up to 3.3.8, highlighting a systemic failure in the plugin's data processing pipeline for comment content."
}