CVAT is an open source interactive video and image annotation tool for computer vision. Versions 2.5.0 through 2.66.0 contain a XSS vulnerability that can be accessed through annotation guide assets. When CVAT serves the files attached to an annotation guide, it labels them with a media type ( Content-Type ) that the attacker can influence, so instead of treating an uploaded file as plain data, the victim's browser can be told to treat it as an HTML page and run any JavaScript inside it. This issue has been fixed in version 2.67.0.
https://github.com/cvat-ai/cvat/security/advisories/GHSA-w6mx-95ff-72cv
https://github.com/cvat-ai/cvat/commit/44d717ad3a9d914f1cb2593ce09efd87d0b9159e
Published: 2026-08-04
Updated: 2026-08-05
Base Score: 4.3
Vector: CVSS2#AV:N/AC:M/Au:N/C:N/I:P/A:N
Severity: Medium
Base Score: 6.1
Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N
Severity: Medium
Base Score: 8.5
Vector: CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:P/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N
Severity: High
EPSS: 0.00239