CVE-2026-69241: Filament Curator: stored SVG XSS in media panel

GHSA-8vm9-f75m-5h2mMEDIUM
Published November 17, 2026
CISO Take

Filament Curator, a media picker plugin for Laravel Filament admin panels, accepts SVG uploads by default and serves them without sanitization, so any user with upload permission can plant a script that runs in the application's origin when another panel user opens the file with the View action. This is a medium-severity issue (CVSS 5.4), but the scope change matters: the script can perform authenticated same-origin actions as the viewer and read script-accessible tokens, so a low-privilege uploader could escalate by targeting an administrator. There is no CISA KEV listing, no public exploit and no Nuclei template, and no EPSS score is available yet, so near-term mass exploitation looks unlikely. The exposure is limited to organizations that use Filament Curator, for example in internal AI/ML dashboards or content portals built on Filament. Upgrade to 3.7.9, 4.1.2 or 5.1.2, and review existing media libraries for SVG files that contain script content.

Sources:NVDGitHub AdvisoryATLAS

What is the risk?

Medium overall. Exploitation is low complexity and needs only an authenticated account with media-upload permission, plus one user interaction (the victim opening the View action). The changed scope lets the payload act with the victim's privileges, so an administrator victim could lead to account takeover or data tampering. No public exploit, KEV listing or scanner template exists. The AI/ML angle is indirect: Filament is a general PHP admin framework, and the risk applies only where it fronts ML tooling, datasets or model administration. Confidence is MEDIUM because package context data is sparse.

How does the attack unfold?

Initial Access
Attacker with media-upload permission uploads a malicious SVG containing script to the Curator media library.
AML.T0049
User Execution
A panel user opens the file via the View action and the SVG runs as a top-level document in the app's origin.
AML.T0011.003
Impact
Script performs authenticated same-origin actions as the victim and reads accessible tokens or data, enabling privilege escalation.

What systems are affected?

PackageEcosystemVulnerable RangePatched
Panelcomposer>= 5.0.0, <= 5.1.15.1.2
5.8KOpenSSF 6.6517 dependentsPushed 3d ago65% patched~15d to patchFull package profile →

Do you use Panel? You're affected.

How severe is it?

CVSS 3.1
5.4 / 10
EPSS
N/A
Exploitation Status
No known exploitation
Sophistication
Trivial

What is the attack surface?

AVACPRUISCIA
AVNetwork
ACLow
PRLow
UIRequired
SChanged
CLow
ILow
ANone

What should I do?

6 steps
  1. Upgrade filament-curator to 3.7.9, 4.1.2 or 5.1.2 depending on your major version.

  2. Until patched, remove image/svg+xml from the accepted MIME types in the Curator configuration, or restrict media-upload permission to trusted users.

  3. Audit existing uploads for SVG files containing <script>, on* event-handler attributes or javascript: URIs, and delete or sanitize them.

  4. Serve uploaded media from a separate cookieless domain, or with Content-Disposition: attachment and a restrictive Content-Security-Policy such as sandbox.

  5. Set session cookies HttpOnly and avoid storing tokens in script-accessible storage.

  6. Detect by reviewing web logs for unusual same-origin requests following media View actions.

How is it classified?

Which compliance frameworks are affected?

This CVE is relevant to:

ISO 42001
A.6.2.6-AI system operation and monitoring
NIST AI RMF
MANAGE 2.3-Mechanisms to respond to and recover from unknown risks

Frequently Asked Questions

What is CVE-2026-69241?

Filament Curator, a media picker plugin for Laravel Filament admin panels, accepts SVG uploads by default and serves them without sanitization, so any user with upload permission can plant a script that runs in the application's origin when another panel user opens the file with the View action. This is a medium-severity issue (CVSS 5.4), but the scope change matters: the script can perform authenticated same-origin actions as the viewer and read script-accessible tokens, so a low-privilege uploader could escalate by targeting an administrator. There is no CISA KEV listing, no public exploit and no Nuclei template, and no EPSS score is available yet, so near-term mass exploitation looks unlikely. The exposure is limited to organizations that use Filament Curator, for example in internal AI/ML dashboards or content portals built on Filament. Upgrade to 3.7.9, 4.1.2 or 5.1.2, and review existing media libraries for SVG files that contain script content.

Is CVE-2026-69241 actively exploited?

No confirmed active exploitation of CVE-2026-69241 has been reported, but organizations should still patch proactively.

How to fix CVE-2026-69241?

1. Upgrade filament-curator to 3.7.9, 4.1.2 or 5.1.2 depending on your major version. 2. Until patched, remove image/svg+xml from the accepted MIME types in the Curator configuration, or restrict media-upload permission to trusted users. 3. Audit existing uploads for SVG files containing <script>, on* event-handler attributes or javascript: URIs, and delete or sanitize them. 4. Serve uploaded media from a separate cookieless domain, or with Content-Disposition: attachment and a restrictive Content-Security-Policy such as sandbox. 5. Set session cookies HttpOnly and avoid storing tokens in script-accessible storage. 6. Detect by reviewing web logs for unusual same-origin requests following media View actions.

What systems are affected by CVE-2026-69241?

This vulnerability affects the following AI/ML architecture patterns: Admin dashboards for ML tooling, Content management panels, Web-based ML UIs.

What is the CVSS score for CVE-2026-69241?

CVE-2026-69241 has a CVSS v3.1 base score of 5.4 (MEDIUM).

What is the AI security impact?

Affected AI Architectures

Admin dashboards for ML toolingContent management panelsWeb-based ML UIs

MITRE ATLAS Techniques

AML.T0011.003Malicious Link
AML.T0049Exploit Public-Facing Application

Compliance Controls Affected

ISO 42001: A.6.2.6
NIST AI RMF: MANAGE 2.3

What are the technical details?

Original Advisory

Filament Curator is a media picker plugin for Filament panels. From 3.0.0 until 3.7.9, 4.1.2, and 5.1.2, the Curator media panel accepts image/svg+xml files by default and stores and serves their markup inline without sanitization. An authenticated user with media-upload permission can store an SVG containing scripts, event-handler attributes, or javascript: URIs. When another panel user selects the View action, the SVG opens as a top-level browser document and its active content executes in the Filament application's origin. The script can perform authenticated same-origin actions as the viewing user and read same-origin data and script-accessible tokens. Panel thumbnails are not affected because they render through an image element that does not execute SVG script. This issue is fixed in versions 3.7.9, 4.1.2, and 5.1.2.

Exploitation Scenario

An attacker with a low-privilege Filament account, such as a contributor or compromised editor, uploads an SVG with an embedded script through the Curator media panel. They then prompt an administrator to open the file with the View action, for example by asking them to review an asset. The SVG opens as a top-level document and the script runs in the panel's origin. It uses the admin's session to create a new privileged user, change settings or read tokens, then sends the data to an attacker-controlled server.

Weaknesses (CWE)

CWE-284 — Improper Access Control: The product does not restrict or incorrectly restricts access to a resource from an unauthorized actor.

  • [Architecture and Design, Operation] Very carefully manage the setting, management, and handling of privileges. Explicitly manage trust zones in the software.
  • [Architecture and Design] Compartmentalize the system to have "safe" areas where trust boundaries can be unambiguously drawn. Do not allow sensitive data to go outside of the trust boundary and always be careful when interfacing with a compartment outside of the safe area. Ensure that appropriate compartmentalization is built into the system design, and the compartmentalization allows for and reinforces privilege separation functionality. Architects and designers should rely on the principle of least privilege to decide the appropriate time to use privileges and the time to drop privileges.

Source: MITRE CWE corpus.

CVSS Vector

CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N

Timeline

Published
November 17, 2026
Last Modified
November 17, 2026
First Seen
November 17, 2026

Related Vulnerabilities