Atomic Edge analysis of CVE-2026-23805:
The Media Search Enhanced WordPress plugin contains an authenticated SQL injection vulnerability. This flaw affects the plugin’s media query filtering functionality, allowing attackers with author-level privileges or higher to execute arbitrary SQL commands. The vulnerability has a CVSS score of 6.5, indicating a medium severity risk.
Atomic Edge research identified the root cause as insufficient input sanitization and a lack of prepared statements for the `post_parent` and `taxonomy` parameters. In the vulnerable file `media-search-enhanced/public/class-media-search-enhanced.php`, the `filter_ajax_query_attachments` function (lines 153-156) directly concatenated user-supplied `$vars[‘post_parent’]` into an SQL `WHERE` clause without validation. Similarly, the `taxonomy` parameter (line 210) was directly interpolated into a SQL string without escaping.
The exploitation vector requires an authenticated attacker with at least Author-level access. Attackers can send a crafted AJAX request to the WordPress `admin-ajax.php` endpoint with the action `query-attachments`. By injecting SQL payloads into the `query[post_parent]` or `query[tax_query][taxonomy]` parameters, they can manipulate the resulting database query. This allows data exfiltration from the WordPress database, including sensitive user information.
The patch in version 0.9.2 addresses the vulnerability by implementing proper input validation and using WordPress database prepared statements. For the `post_parent` parameter, the code now uses `absint()` to cast the value to a non-negative integer and `$wpdb->prepare()` with a `%d` placeholder (lines 154-160). For the `taxonomy` parameter, the patch adds `sanitize_key()` and uses `$wpdb->prepare()` with a `%s` placeholder (lines 210-211). These changes ensure user input is properly sanitized before inclusion in SQL queries.
Successful exploitation enables unauthorized data extraction from the WordPress database. Attackers can retrieve sensitive information such as password hashes, user emails, and private post content. This can facilitate further attacks, including site compromise and privilege escalation. The requirement for author-level authentication limits immediate attack surface but poses a significant risk in multi-user environments.







