justhtml before version 1.15.0 contains multiple security flaws related to improper input validation in URL sanitization helpers (clean_url_value, clean_url_in_js_string), HTML serialization, Markdown passthrough when html_passthrough=True, and several edge cases in custom sanitization policies. These flaws allow attackers to bypass sanitization mechanisms and inject active HTML and JavaScript code. Examples include encoded javascript: URLs, backslash-based relative URLs resolved as remote hosts, markup-breaking element or attribute names, HTML comments, raw </textarea> tags reintroduced via Markdown passthrough, and preservation of <style>, <meta http-equiv=refresh>, and <base href> tags in custom policies. The default sanitize=true configuration is mostly unaffected; the issues primarily impact advanced usage scenarios involving helper APIs, programmatic DOM construction, html_passthrough=True, and custom sanitization policies or transformation pipelines.

An attacker can bypass input sanitization to inject active HTML and JavaScript code, potentially leading to cross-site scripting (XSS) or other injection-based attacks. This can compromise the security of applications using vulnerable justhtml versions with certain configurations, especially those using html_passthrough=True or custom sanitization policies. The vulnerability is critical, indicating a high potential impact on confidentiality, integrity, and availability if exploited.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until an official fix is available, avoid using html_passthrough=True and custom sanitization policies that may be vulnerable. Prefer the default sanitize=true configuration, which is mostly unaffected by these issues. Monitor the vendor's communications for updates and apply patches promptly once released.