---

Overview

A new vulnerability, designated CVE-2025-22270, was discovered in CyberArk Endpoint Privilege Manager SaaS version 24.7.1. This security issue resides in the "Role Management" tab of the Administration panel. Attackers who can access this panel can inject malicious code through the "name" field when creating a new role.

However, there’s an important twist: because of the way modern browsers and web apps are secured, exploiting this vulnerability to run JavaScript (and thus do more harm) is tough. The web app has a Content Security Policy (CSP) in place, and unless something else goes wrong (like a CSP bypass), you’re looking at an HTML injection rather than full-blown JavaScript execution.

References

- Original CVE Entry (Coming Soon)
- CyberArk Endpoint Privilege Manager

How the Attack Works

This vulnerability can only be exploited by users with access to the Administration panel of CyberArk Endpoint Privilege Manager. Here’s a step-by-step overview:

Attacker Logs In: The attacker must have valid credentials and access rights.

2. Navigates to Role Management: In the “Administration” section, under “Role Management,” the attacker starts to create a new role.
3. Injects Malicious Code in Name Field: Instead of a normal name, the attacker submits special HTML/JavaScript.

Example Payload

<script>alert('CVE-2025-22270')</script>

or, for a proof-of-concept,

<img src="x" onerror="alert('HTML Injection!')"/>

4. Payload Stored: The server does not sanitize input, stores the code, and displays it elsewhere in the application without proper escaping.

What Are the Dangers?

Because of the existing Content Security Policy (CSP), the application blocks much of what an attacker could do with standard JavaScript injection (like stealing cookies or session tokens). However, HTML injection is still possible, so an attacker could:

- Mess with the look/feel of the admin interface (defacement).

If CSP can be bypassed by another error, escalate to full JavaScript injection.

It’s important to note: This is not a "public exploit" risk for random attackers across the Internet. Attackers need valid admin access to the panel—so this is a privilege escalation risk within an already-compromised environment.

Exploit Example

Here’s a simple exploit demonstration. The attacker creates a role and fills the role name with HTML code:

New Role Name: <img src="x" onerror="alert('You have CVE-2025-22270!')"/>

When someone later views the roles list, they’ll see

![](https://user-images.githubusercontent.com/PLACEHOLDER/role-injection-demo.png)

And the browser will pop up an alert (or whatever the attacker wants to put there).

Scope Limited: Only admin users can exploit this (it doesn’t allow outside hackers in).

- CSP Mitigates Impact: The Content Security Policy greatly reduces the chance for JavaScript execution (XSS); only HTML injection is possible unless there’s another bug.
- Potential for Chaining: If another vulnerability lets the attacker bypass CSP, things could get worse quickly.

Vendor Response

Despite multiple attempts to contact CyberArk, the vendor did not respond to responsible disclosure notifications. There is no patch or mitigation information available at the time of writing.

If you use CyberArk Endpoint Privilege Manager SaaS version 24.7.1:

Conclusion

CVE-2025-22270 shows how even "small" vulnerabilities can matter. While the main risk for now is limited to visual tampering or mild phishing, the underlying lack of sanitization could become more dangerous if paired with a CSP bypass.

For more details

- CyberArk Endpoint Privilege Manager
- Stay tuned for the CVE page

---
*Exclusive for you by Security Insights Team – Security news made simple.*

Timeline

Published on: 02/28/2025 13:15:27 UTC
Last modified on: 03/05/2025 16:15:37 UTC