Cross-site Scripting in pkp/pkp-lib

Successful exploitation of CVE-2023-5901 allows an attacker to execute arbitrary JavaScript code within the context of a victim's browser session. This can be leveraged to steal sensitive information, such as cookies and session tokens, granting the attacker unauthorized access to the user's account. The attacker could also deface the website, redirect users to malicious sites, or perform other actions on behalf of the compromised user. The impact is particularly concerning for OJS installations used by academic institutions and publishers, as these often handle sensitive user data and intellectual property.

Organizations and institutions running Open Journal Systems (OJS) with versions of pkp-lib prior to 3.3.0-16 are at risk. This includes academic publishers, university libraries, and research institutions that rely on OJS for managing their journals and publications. Shared hosting environments using OJS are particularly vulnerable due to the potential for cross-tenant contamination.

• php / web: Examine OJS application logs for suspicious JavaScript injection attempts. Use a web application firewall (WAF) to filter out common XSS payloads. • generic web: Use curl/wget to test form fields for XSS vulnerabilities. Inspect response headers for unexpected script tags.

curl -X POST -d "<script>alert(1)</script>" http://your-ojs-instance/index.php/your-journal/about

1. 2023-11-01Disclosure

   disclosure

1. आरक्षित2023-11-01
2. प्रकाशित2023-11-01
3. संशोधित2024-08-02
4. EPSS अद्यतन2026-04-02

The primary mitigation for CVE-2023-5901 is to upgrade pkp-lib to version 3.3.0-16 or later. If an immediate upgrade is not possible due to compatibility issues or downtime constraints, consider implementing input validation and output encoding on user-supplied data within your OJS installation. While not a complete solution, this can help reduce the attack surface. Regularly review and update your OJS installation to ensure you are running the latest security patches. After upgrading, confirm the fix by attempting to inject a simple XSS payload (e.g., <script>alert(1)</script>) into a form field and verifying that the script is not executed.