Gitea's Jupyter notebook renderer inserts the notebook's language_info.name value into a CSS class attribute without sanitizing it or using a sandboxed frame. A malicious notebook can supply multiple CSS classes that create a viewport‑covering overlay and capture user clicks, which causes UI redress. No arbitrary script execution was demonstrated; the impact is purely visual and interaction hijacking.

All Gitea installations that expose the Jupyter notebook rendering page potentially risk this vulnerability. The available data does not specify a fixed product version, so any deployment that includes this feature is at risk until a fix is applied.

The CVSS score of 3.5 indicates a low severity impact. EPSS score is < 1% and the vulnerability is not listed in the CISA KEV catalog. Exploitation requires an attacker to upload a malicious notebook to a repository that another user will view, and the victim must use a browser to see the notebook. Because no remote code execution was demonstrated and the attack surface is limited to in-browser visual manipulation, the overall threat remains moderate.