CVE-2026-49825: lxml-html-clean - XSS Vulnerability Due to Incomplete URL Scheme Filtering

Sponsored Habilume Habit tracker with a companion that grows with you Complete habits, earn Lume Points, and watch your Lume grow as you stay consistent. App Store Google Play

Background

lxml-html-clean is a Python package that provides a cleaner for HTML content, allowing developers to sanitize untrusted HTML input. It is commonly used to prevent cross-site scripting (XSS) attacks by filtering out potentially dangerous elements and attributes. However, certain configurations of the Cleaner class can lead to vulnerabilities if not properly managed, particularly with namespaced URL attributes.

Vulnerability Detail

The vulnerability arises from the lxml_html_clean.Cleaner not stripping 'javascript:' URLs from namespaced URL attributes such as 'xlink:href'. This issue is rooted in the implementation of the Cleaner, which relies on a predefined allow-list of attributes that does not include namespaced attributes. As a result, when the Cleaner is configured with 'safe_attrs_only=False', malicious payloads can pass through the sanitizer, leading to stored XSS vulnerabilities. This flaw can be exploited when a user clicks on a rendered link, executing the embedded JavaScript in the context of the application.

pip install --upgrade "lxml-html-clean==0.4.5"

Workaround

As an immediate workaround, you can manually extend the Cleaner class to include a check for namespaced URL attributes. Here is a code snippet to implement this workaround:

from lxml import html
from lxml_html_clean import Cleaner
import re

class CustomCleaner(Cleaner):
    def __call__(self, doc):
        super().__call__(doc)
        for el in doc.iter():
            if 'xlink:href' in el.attrib:
                if re.match(r'^\s*(javascript|vbscript|data):', el.attrib['xlink:href'], re.I):
                    del el.attrib['xlink:href']

This custom cleaner will remove any 'javascript:' URLs from 'xlink:href' attributes.

Conclusion

The CVE-2026-49825 vulnerability in lxml-html-clean poses a significant risk of XSS attacks due to improper handling of namespaced URL attributes. It is crucial for developers to update to the patched version or implement a workaround to mitigate this risk.

See every vulnerability recorded for lxml-html-clean →

FAQ