NoteGen vulnerabilities

NoteGen through 0.38.3 contains an external control of file path vulnerability in chat image analysis that allows attackers to read local image files via unvalidated synced sourceUrl paths. Attackers with write access to the victim's sync target can plant absolute paths or file:// URLs, causing image_inspect to send local images to the AI provider.

NoteGen before version 0.32.0 grants the Tauri shell plugin shell:allow-execute capability for bash, python, and python3 with arbitrary arguments by default. JavaScript in the application webview can invoke plugin:shell|execute to run attacker-controlled OS commands with NoteGen process privileges.

NoteGen before version 0.32.0 renders AI chat responses with markdown-it configured with html:true and injects the result into the DOM via dangerouslySetInnerHTML without HTML sanitization and with CSP set to null. Attacker-controlled content can cause the model response to include executable markup, leading to XSS in the application context.