CVE-2025-31564: Ai Auto Tool Plugin SQLi Vulnerability

CVE-2025-31564 Overview

CVE-2025-31564 is a Blind SQL Injection vulnerability in the WordPress plugin ai-auto-tool (Ai Auto Tool Content Writing Assistant - Gemini Writer, ChatGPT All in One). The flaw stems from improper neutralization of special elements used in SQL commands [CWE-89]. All plugin versions up to and including 2.2.6 are affected. An authenticated attacker with low privileges can inject SQL fragments through unsanitized parameters and infer database contents through boolean or time-based responses. The vulnerability has a scope change impact, meaning successful exploitation can affect resources beyond the vulnerable component.

Critical Impact
Authenticated attackers can extract sensitive WordPress database contents, including user credentials and session data, through blind SQL injection against the affected plugin.

Affected Products

Ai Auto Tool Content Writing Assistant (Gemini Writer, ChatGPT) All in One WordPress plugin
Plugin slug: ai-auto-tool
All versions from n/a through 2.2.6

Discovery Timeline

2025-04-01 - CVE-2025-31564 published to NVD
2026-06-17 - Last updated in NVD database

Technical Details for CVE-2025-31564

Vulnerability Analysis

The vulnerability resides in the ai-auto-tool WordPress plugin, which integrates Gemini and ChatGPT content generation features. User-controlled input passes into SQL statements without proper sanitization or parameterization. Because the application does not return query results directly, exploitation follows the Blind SQL Injection pattern, where the attacker infers data through application response differences or response timing.

The Patchstack advisory categorizes this as a Blind SQL Injection issue affecting all versions through 2.2.6. Exploitation requires network access and low-privileged authentication, with no user interaction needed.

Root Cause

The root cause is the absence of prepared statements or proper escaping when constructing SQL queries from request parameters. WordPress provides the $wpdb->prepare() API for parameterized queries, but the plugin concatenates untrusted input directly into queries. This pattern matches [CWE-89] — Improper Neutralization of Special Elements used in an SQL Command.

Attack Vector

An attacker authenticates to the target WordPress site with a low-privileged account, such as Subscriber or Contributor, depending on the affected endpoint. The attacker then submits crafted requests to the plugin's vulnerable parameter, embedding SQL payloads that alter query logic. Boolean-based payloads compare conditions and observe page response differences. Time-based payloads use functions like SLEEP() to infer data through response delays. Repeated requests allow character-by-character extraction of database contents, including the wp_users table.

No verified public proof-of-concept code is available at the time of writing. Refer to the Patchstack SQL Injection Vulnerability advisory for additional technical context.

Detection Methods for CVE-2025-31564

Indicators of Compromise

HTTP requests to ai-auto-tool plugin endpoints containing SQL keywords such as SELECT, UNION, SLEEP, BENCHMARK, or INFORMATION_SCHEMA
Repeated parameter requests with incrementing numeric or character values consistent with blind extraction
WordPress access logs showing low-privileged authenticated users issuing high-frequency requests to plugin AJAX or REST endpoints
Database error entries or unusually long query execution times in MySQL slow query logs

Detection Strategies

Inspect web server logs for query strings or POST bodies referencing the ai-auto-tool plugin path together with SQL meta-characters such as ', --, ;, or /*
Deploy a Web Application Firewall ruleset that flags SQL injection signatures against WordPress plugin endpoints
Monitor database query patterns for time-delay functions and unusual subqueries originating from the WordPress process

Monitoring Recommendations

Enable MySQL general or slow query logging and forward logs to a centralized analytics platform for correlation
Alert on authenticated WordPress sessions that generate elevated error rates or response-time variance on plugin endpoints
Track plugin version inventory across WordPress estates to identify hosts still running ai-auto-tool 2.2.6 or earlier

How to Mitigate CVE-2025-31564

Immediate Actions Required

Identify all WordPress sites running the ai-auto-tool plugin at version 2.2.6 or earlier
Deactivate and remove the plugin until a patched release is verified and deployed
Rotate WordPress administrator credentials and database passwords if exploitation is suspected
Audit the wp_users and wp_usermeta tables for unauthorized accounts or modified roles

Patch Information

At the time of NVD publication, the vendor advisory referenced through Patchstack lists all versions through 2.2.6 as affected. Administrators should consult the WordPress.org plugin directory and the Patchstack listing for current patched versions and apply updates as soon as they are released.

Workarounds

Disable the ai-auto-tool plugin if a patched version is not yet available
Restrict access to WordPress authenticated endpoints behind a Web Application Firewall configured with SQL injection rules
Apply the principle of least privilege to WordPress user roles and remove unused low-privileged accounts
Enforce strong authentication, including multi-factor authentication, on all WordPress accounts to limit abuse of low-privileged credentials

bash

# Example: disable the vulnerable plugin via WP-CLI
wp plugin deactivate ai-auto-tool
wp plugin delete ai-auto-tool