The vulnerability CVE-2026-32940 affects SiYuan, a personal knowledge management system, in versions prior to 3.6.1. It stems from an incomplete sanitization mechanism in the SanitizeSVG function, which attempts to block dangerous data URI schemes in href attributes within SVG content but overlooks data:text/xml and data:application/xml. These overlooked schemes can embed SVG images containing JavaScript, enabling script execution. The unauthenticated API endpoint /api/icon/getDynamicIcon accepts a content parameter that is directly inserted into SVG markup using Go's fmt.Sprintf without any escaping or validation, and the response is served with the MIME type image/svg+xml. This creates a click-through cross-site scripting (XSS) vulnerability: an attacker can craft a URL that returns an SVG containing a malicious link. When a victim navigates to this URL or the SVG is embedded via <object> or <embed> tags, clicking the link triggers JavaScript execution in the victim's browser context. The vulnerability requires user interaction (click) and direct navigation or embedding, as <img> tags do not support interactive links in SVG. This flaw compromises confidentiality and integrity by enabling script injection and potential session hijacking or data theft. The vulnerability has been addressed in SiYuan version 3.6.1 by improving the sanitization logic to block the bypassed data URI schemes. No known exploits are reported in the wild as of now. The CVSS v3.1 score of 9.3 indicates a critical severity due to network attack vector, no privileges required, low attack complexity, requirement for user interaction, and a scope change affecting confidentiality and integrity.

This vulnerability poses a significant risk to organizations using SiYuan versions 3.6.0 and below, especially those deploying it in environments accessible over the internet. Successful exploitation allows attackers to execute arbitrary JavaScript in the context of the victim's browser, potentially leading to session hijacking, theft of sensitive information, unauthorized actions on behalf of the user, or delivery of further malware. Since the vulnerable endpoint is unauthenticated and accessible remotely, attackers can craft malicious URLs that can be distributed via phishing or social engineering campaigns. The requirement for user interaction (click) somewhat limits automated exploitation but does not diminish the risk in targeted attacks. Organizations relying on SiYuan for knowledge management may face data confidentiality breaches and integrity violations. The vulnerability does not affect availability directly but can undermine trust and lead to indirect service disruptions. Given the critical CVSS score and the nature of the vulnerability, it demands urgent remediation to prevent exploitation.

Organizations should immediately upgrade SiYuan to version 3.6.1 or later, where the vulnerability is fixed. Until upgrade is possible, implement strict network controls to restrict access to the /api/icon/getDynamicIcon endpoint, limiting it to trusted users or internal networks only. Employ web application firewalls (WAFs) with custom rules to detect and block requests containing suspicious data URI schemes in SVG content parameters. Educate users about the risks of clicking unknown or suspicious links, especially those pointing to SVG resources. Review and sanitize all user-supplied inputs that may be rendered as SVG or other markup formats, applying a whitelist approach rather than blacklists. Monitor logs for unusual access patterns to the vulnerable endpoint and signs of attempted exploitation. Consider disabling or restricting SVG rendering features if not essential. Finally, conduct security assessments and penetration tests focusing on SVG and other vector graphic handling in web applications.