GCVE Workshop - 22 September 2026 (14:00-18:00), Luxembourg Before The Vulnopticon Conference - Registration

GHSA-4JHM-JV67-739F

Vulnerability from github – Published: 2026-07-08 20:23 – Updated: 2026-07-08 20:23
VLAI
Summary
`lxml_html_clean.Cleaner` does not strip `javascript:` URLs from namespaced URL attributes
Details

lxml_html_clean.Cleaner does not strip javascript: URLs from namespaced URL attributes (xlink:href)

Reporter: Guillem Lefait guillem@datamq.com · Date: 2026-05-10 Affected: lxml ≤ 6.1.0 and lxml_html_clean ≤ 0.4.4 (latest stable) Confirmed against: lxml 6.1.0 + lxml_html_clean 0.4.4 on Python 3.13.5, 3.14.4, and 3.15.0a8 (libxml2 2.14.6 / 2.9.14 — bug is in pure-Python sanitizer logic, independent of the libxml2 backend) Root-cause class: same as CVE-2021-28957 (formaction missing from link_attrs)

Summary

Cleaner filters URL schemes (javascript:, vbscript:, …) by walking links via rewrite_links(), which delegates to iterlinks(), which only yields attributes named in lxml.html.defs.link_attrs. That allow-list contains no prefixed names (xlink:href) and no srcset. As a result, when Cleaner is configured with safe_attrs_only=False — a documented option for callers that want lenient attribute handling but still expect URL-scheme scrubbing — <a xlink:href="javascript:…"> survives sanitization untouched, and any browser that follows the SVG-anchor specification will execute the JavaScript when the rendered link is clicked.

CWE: CWE-79 (XSS), with CWE-184 (Incomplete List of Disallowed Inputs) as the underlying defect class.

Affected components

Package Versions tested File / line
lxml 4.9.x, 5.2.1, 6.1.0 src/lxml/html/defs.py:20
lxml " src/lxml/html/__init__.py:485-528
lxml_html_clean 0.4.0 – 0.4.4 lxml_html_clean/clean.py:348,576

The legacy lxml.html.clean module — bundled in lxml < 5.2.0 and still installable on newer versions via the lxml[html_clean] extra — shares the same bug.

Root cause

defs.link_attrs is a flat string set; the literal xlink:href is absent:

# lxml/html/defs.py
link_attrs = frozenset([
    'action', 'archive', 'background', 'cite', 'classid',
    'codebase', 'data', 'href', 'longdesc', 'profile', 'src',
    'usemap', 'dynsrc', 'lowsrc', 'formaction',
])

HtmlMixin.iterlinks() (lxml/html/__init__.py:526-528) only yields attributes whose key is in that set:

for attrib in link_attrs:
    if attrib in attribs:
        yield (el, attrib, attribs[attrib], 0)

Cleaner.__call__ registers the URL-scheme filter via rewrite_links (lxml_html_clean/clean.py:348), which is a thin wrapper around iterlinks(). Because xlink:href is never yielded, _remove_javascript_link (clean.py:576) is never invoked for it.

Minimal reproducer

from lxml import html
from lxml_html_clean import Cleaner

for payload in (
    '<svg><a xlink:href="javascript:alert(1)">x</a></svg>',
    '<math><a xlink:href="javascript:alert(2)">y</a></math>',
):
    tree = html.fromstring(payload)
    Cleaner(safe_attrs_only=False)(tree)
    print(html.tostring(tree).decode())
    print('  iterlinks:', list(html.fromstring(payload).iterlinks()))
# <svg><a xlink:href="javascript:alert(1)">x</a></svg>     ← unchanged
#   iterlinks: []                                          ← link rewriter blind
# <math><a xlink:href="javascript:alert(2)">y</a></math>   ← unchanged
#   iterlinks: []                                          ← link rewriter blind

Both SVG and MathML scopes are vulnerable — same allow-list gap, both render anchors that browsers treat as navigable. Other lab-confirmed surviving variants (same scope, different scheme encoding): mixed-case (JaVaScRiPt:), HTML-entity (java&#x73;cript:), embedded tab (java\tscript:).

Impact

A caller that uses Cleaner to neutralise untrusted HTML and chooses safe_attrs_only=False — typically because the application wants to allow custom data-/aria-/vendor attributes — will silently pass javascript: payloads carried on xlink:href through to victim renders. Stored XSS in any application that round-trips user-supplied HTML through this configuration. Reach is conditional on the safe_attrs_only=False toggle, but that is a documented public option; consumers reasonably expect URL-scheme scrubbing to be independent of attribute allow-listing.

Suggested fix

Extend link_attrs to include xlink:href. In HTML mode, lxml.html keeps prefixed attribute names verbatim — the parsed key is the literal string xlink:href, not a Clark-notation form — so the existing allow-list lookup is a plain string match. Same shape as the CVE-2021-28957 fix:

 # lxml/html/defs.py
 link_attrs = frozenset([
     'action', 'archive', 'background', 'cite', 'classid',
     'codebase', 'data', 'href', 'longdesc', 'profile', 'src',
     'usemap', 'dynsrc', 'lowsrc', 'formaction',
+    'xlink:href',
 ])

This single change closes the reported XSS for both SVG <a xlink:href> and MathML <a xlink:href>. lxml_html_clean is the canonical home of the Cleaner code (881 lines); lxml.html.clean is a 21-line backward-compat shim (from lxml_html_clean import *) that picks up the fix automatically once link_attrs is updated upstream. Since the upstream change requires lxml maintainer action, see the alternative below if a self-contained patch in lxml_html_clean is preferred.

Alternative (in-package fix, no lxml coordination needed): add a namespaced-URL-attribute walk inside Cleaner.__call__ so the URL-scheme filter doesn't depend on link_attrs. Sketch:

# lxml_html_clean/clean.py — supplements rewrite_links() in __call__
_NS_URL_ATTRS = ('xlink:href',)  # extend as needed
_BAD_SCHEME = re.compile(r'^\s*(javascript|vbscript|data):', re.I)

for el in doc.iter():
    for attr in _NS_URL_ATTRS:
        if attr in el.attrib and _BAD_SCHEME.match(el.attrib[attr]):
            del el.attrib[attr]

This decouples the cleaner from the upstream link_attrs set and matches the security-ownership boundary established when the cleaner was extracted in lxml 5.2.0.

Defense in depth (optional, regardless of which fix path is taken): - Also handle srcset: the value is a url 1x, url 2x, … descriptor list, so split on commas and validate each candidate URL. Not directly executable in current browsers, but closes the same gap. - Also accept Clark-notation forms ({http://www.w3.org/1999/xlink}href) so XML-mode callers using lxml.etree get the same protection. HTML mode never produces this form, so not needed for the reported bug.

Severity

CVSS 3.1 base score: 8.2 / High — AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:L/A:N (stored XSS; victim must click the SVG anchor; scope-changed because script executes in the rendering origin). PR:N reflects the common case where untrusted HTML enters the sanitizer from anonymous sources (comments, support tickets); deployments that gate writes behind authentication can score with PR:L (→ 7.6).

Severity is CONDITIONAL on the caller passing safe_attrs_only=False. With the class default (True), attribute allow-listing strips xlink:href before scheme scrubbing runs, and the bug does not fire — verified at HEAD: default-config Cleaner()(<svg><a xlink:href="javascript:…">x</a></svg>) → <svg><a>x</a></svg>.

Prior art / novelty

  • CVE-2021-28957 (lxml 4.6.3) — same root cause, different attribute (formaction). Fix was a one-line extension of link_attrs. Direct precedent.
  • CVE-2022-34473 (Mozilla Sanitizer API) — xlink:href URL bypass primitive in a different sanitizer.
  • Bleach (Mozilla, Python) explicitly handles the xlink namespace; enshrined/svg-sanitize (PHP) ships cleanXlinkHrefs(); DOMPurify scrubs xlink:href via ALLOWED_URI_REGEXP.
  • nh3 (the alternative recommended in lxml_html_clean's own README for security-sensitive use) is not vulnerable to this primitive — verified 2026-05-10 on nh3==0.3.5: with <svg>/<math>/<a> and xlink:href explicitly added to tags/attributes, both SVG and MathML payloads, all four scheme-encoding variants, are stripped (output e.g. <svg><a rel="noopener noreferrer">x</a></svg>).

Coordination

Filing as a private GHSA at fedora-python/lxml_html_clean — lxml_html_clean is the canonical maintainer of the Cleaner code (881 lines) and the security-responsible team since the lxml 5.2.0 split, where the cleaner was extracted out of lxml precisely so cleaner-security reports could land on the right team. The lxml side cannot be filed via GHSA (https://github.com/lxml/lxml/security/advisories/new returns 404 — private reporting is not enabled), so a parallel report has been emailed directly to the lxml maintainer for the upstream defs.link_attrs patch path. You're welcome to coordinate with them directly if you'd prefer the upstream fix over the in-package alternative above.

Happy to provide a draft patch or PR on either path. No bounty expected.

Severity
8.2 (High)
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:L/A:N
Show details on source website
To clipboard 

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "lxml_html_clean"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.4.5"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-49825"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-184",
      "CWE-79"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-07-08T20:23:22Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "# `lxml_html_clean.Cleaner` does not strip `javascript:` URLs from namespaced URL attributes (`xlink:href`)\n\n**Reporter:** Guillem Lefait \u003cguillem@datamq.com\u003e \u00b7 **Date:** 2026-05-10\n**Affected:** `lxml` \u2264 6.1.0 and `lxml_html_clean` \u2264 0.4.4 (latest stable)\n**Confirmed against:** lxml 6.1.0 + lxml_html_clean 0.4.4 on Python 3.13.5, 3.14.4, and 3.15.0a8 (libxml2 2.14.6 / 2.9.14 \u2014 bug is in pure-Python sanitizer logic, independent of the libxml2 backend)\n**Root-cause class:** same as CVE-2021-28957 (`formaction` missing from `link_attrs`)\n\n## Summary\n\n`Cleaner` filters URL schemes (`javascript:`, `vbscript:`, \u2026) by walking links via `rewrite_links()`, which delegates to `iterlinks()`, which only yields attributes named in `lxml.html.defs.link_attrs`. That allow-list contains no prefixed names (`xlink:href`) and no `srcset`. As a result, when `Cleaner` is configured with `safe_attrs_only=False` \u2014 a documented option for callers that want lenient attribute handling but still expect URL-scheme scrubbing \u2014 `\u003ca xlink:href=\"javascript:\u2026\"\u003e` survives sanitization untouched, and any browser that follows the SVG-anchor specification will execute the JavaScript when the rendered link is clicked.\n\n**CWE:** CWE-79 (XSS), with CWE-184 (Incomplete List of Disallowed Inputs) as the underlying defect class.\n\n## Affected components\n\n| Package            | Versions tested        | File / line                      |\n|--------------------|------------------------|----------------------------------|\n| `lxml`             | 4.9.x, 5.2.1, 6.1.0    | `src/lxml/html/defs.py:20`       |\n| `lxml`             | \"                      | `src/lxml/html/__init__.py:485-528` |\n| `lxml_html_clean`  | 0.4.0 \u2013 0.4.4          | `lxml_html_clean/clean.py:348,576` |\n\nThe legacy `lxml.html.clean` module \u2014 bundled in `lxml \u003c 5.2.0` and still installable on newer versions via the `lxml[html_clean]` extra \u2014 shares the same bug.\n\n## Root cause\n\n`defs.link_attrs` is a flat string set; the literal `xlink:href` is absent:\n\n```python\n# lxml/html/defs.py\nlink_attrs = frozenset([\n    \u0027action\u0027, \u0027archive\u0027, \u0027background\u0027, \u0027cite\u0027, \u0027classid\u0027,\n    \u0027codebase\u0027, \u0027data\u0027, \u0027href\u0027, \u0027longdesc\u0027, \u0027profile\u0027, \u0027src\u0027,\n    \u0027usemap\u0027, \u0027dynsrc\u0027, \u0027lowsrc\u0027, \u0027formaction\u0027,\n])\n```\n\n`HtmlMixin.iterlinks()` (`lxml/html/__init__.py:526-528`) only yields attributes whose key is in that set:\n\n```python\nfor attrib in link_attrs:\n    if attrib in attribs:\n        yield (el, attrib, attribs[attrib], 0)\n```\n\n`Cleaner.__call__` registers the URL-scheme filter via `rewrite_links` (`lxml_html_clean/clean.py:348`), which is a thin wrapper around `iterlinks()`. Because `xlink:href` is never yielded, `_remove_javascript_link` (`clean.py:576`) is never invoked for it.\n\n## Minimal reproducer\n\n```python\nfrom lxml import html\nfrom lxml_html_clean import Cleaner\n\nfor payload in (\n    \u0027\u003csvg\u003e\u003ca xlink:href=\"javascript:alert(1)\"\u003ex\u003c/a\u003e\u003c/svg\u003e\u0027,\n    \u0027\u003cmath\u003e\u003ca xlink:href=\"javascript:alert(2)\"\u003ey\u003c/a\u003e\u003c/math\u003e\u0027,\n):\n    tree = html.fromstring(payload)\n    Cleaner(safe_attrs_only=False)(tree)\n    print(html.tostring(tree).decode())\n    print(\u0027  iterlinks:\u0027, list(html.fromstring(payload).iterlinks()))\n# \u003csvg\u003e\u003ca xlink:href=\"javascript:alert(1)\"\u003ex\u003c/a\u003e\u003c/svg\u003e     \u2190 unchanged\n#   iterlinks: []                                          \u2190 link rewriter blind\n# \u003cmath\u003e\u003ca xlink:href=\"javascript:alert(2)\"\u003ey\u003c/a\u003e\u003c/math\u003e   \u2190 unchanged\n#   iterlinks: []                                          \u2190 link rewriter blind\n```\n\nBoth SVG and MathML scopes are vulnerable \u2014 same allow-list gap, both render anchors that browsers treat as navigable. Other lab-confirmed surviving variants (same scope, different scheme encoding): mixed-case (`JaVaScRiPt:`), HTML-entity (`java\u0026#x73;cript:`), embedded tab (`java\\tscript:`).\n\n## Impact\n\nA caller that uses `Cleaner` to neutralise untrusted HTML and chooses `safe_attrs_only=False` \u2014 typically because the application wants to allow custom data-/aria-/vendor attributes \u2014 will silently pass `javascript:` payloads carried on `xlink:href` through to victim renders. Stored XSS in any application that round-trips user-supplied HTML through this configuration. Reach is conditional on the `safe_attrs_only=False` toggle, but that is a documented public option; consumers reasonably expect URL-scheme scrubbing to be independent of attribute allow-listing.\n\n## Suggested fix\n\n**Extend `link_attrs`** to include `xlink:href`. In HTML mode, `lxml.html` keeps prefixed attribute names verbatim \u2014 the parsed key is the literal string `xlink:href`, not a Clark-notation form \u2014 so the existing allow-list lookup is a plain string match. Same shape as the CVE-2021-28957 fix:\n\n```diff\n # lxml/html/defs.py\n link_attrs = frozenset([\n     \u0027action\u0027, \u0027archive\u0027, \u0027background\u0027, \u0027cite\u0027, \u0027classid\u0027,\n     \u0027codebase\u0027, \u0027data\u0027, \u0027href\u0027, \u0027longdesc\u0027, \u0027profile\u0027, \u0027src\u0027,\n     \u0027usemap\u0027, \u0027dynsrc\u0027, \u0027lowsrc\u0027, \u0027formaction\u0027,\n+    \u0027xlink:href\u0027,\n ])\n```\n\nThis single change closes the reported XSS for both SVG `\u003ca xlink:href\u003e` and MathML `\u003ca xlink:href\u003e`. `lxml_html_clean` is the canonical home of the `Cleaner` code (881 lines); `lxml.html.clean` is a 21-line backward-compat shim (`from lxml_html_clean import *`) that picks up the fix automatically once `link_attrs` is updated upstream. Since the upstream change requires lxml maintainer action, see the alternative below if a self-contained patch in `lxml_html_clean` is preferred.\n\n**Alternative (in-package fix, no lxml coordination needed):** add a namespaced-URL-attribute walk inside `Cleaner.__call__` so the URL-scheme filter doesn\u0027t depend on `link_attrs`. Sketch:\n\n```python\n# lxml_html_clean/clean.py \u2014 supplements rewrite_links() in __call__\n_NS_URL_ATTRS = (\u0027xlink:href\u0027,)  # extend as needed\n_BAD_SCHEME = re.compile(r\u0027^\\s*(javascript|vbscript|data):\u0027, re.I)\n\nfor el in doc.iter():\n    for attr in _NS_URL_ATTRS:\n        if attr in el.attrib and _BAD_SCHEME.match(el.attrib[attr]):\n            del el.attrib[attr]\n```\n\nThis decouples the cleaner from the upstream `link_attrs` set and matches the security-ownership boundary established when the cleaner was extracted in lxml 5.2.0.\n\n**Defense in depth (optional, regardless of which fix path is taken):**\n- Also handle `srcset`: the value is a `url 1x, url 2x, \u2026` descriptor list, so split on commas and validate each candidate URL. Not directly executable in current browsers, but closes the same gap.\n- Also accept Clark-notation forms (`{http://www.w3.org/1999/xlink}href`) so XML-mode callers using `lxml.etree` get the same protection. HTML mode never produces this form, so not needed for the reported bug.\n\n## Severity\n\nCVSS 3.1 base score: **8.2 / High** \u2014 `AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:L/A:N` (stored XSS; victim must click the SVG anchor; scope-changed because script executes in the rendering origin). PR:N reflects the common case where untrusted HTML enters the sanitizer from anonymous sources (comments, support tickets); deployments that gate writes behind authentication can score with PR:L (\u2192 7.6).\n\nSeverity is **CONDITIONAL** on the caller passing `safe_attrs_only=False`. With the class default (`True`), attribute allow-listing strips `xlink:href` before scheme scrubbing runs, and the bug does not fire \u2014 verified at HEAD: default-config `Cleaner()(\u003csvg\u003e\u003ca xlink:href=\"javascript:\u2026\"\u003ex\u003c/a\u003e\u003c/svg\u003e)` \u2192 `\u003csvg\u003e\u003ca\u003ex\u003c/a\u003e\u003c/svg\u003e`.\n\n## Prior art / novelty\n\n- **CVE-2021-28957 (lxml 4.6.3)** \u2014 same root cause, different attribute (`formaction`). Fix was a one-line extension of `link_attrs`. Direct precedent.\n- **CVE-2022-34473** (Mozilla Sanitizer API) \u2014 `xlink:href` URL bypass primitive in a different sanitizer.\n- **Bleach (Mozilla, Python)** explicitly handles the `xlink` namespace; `enshrined/svg-sanitize` (PHP) ships `cleanXlinkHrefs()`; DOMPurify scrubs `xlink:href` via `ALLOWED_URI_REGEXP`.\n- **`nh3`** (the alternative recommended in `lxml_html_clean`\u0027s own README for security-sensitive use) is **not vulnerable** to this primitive \u2014 verified 2026-05-10 on `nh3==0.3.5`: with `\u003csvg\u003e`/`\u003cmath\u003e`/`\u003ca\u003e` and `xlink:href` explicitly added to `tags`/`attributes`, both SVG and MathML payloads, all four scheme-encoding variants, are stripped (output e.g. `\u003csvg\u003e\u003ca rel=\"noopener noreferrer\"\u003ex\u003c/a\u003e\u003c/svg\u003e`).\n\n\n## Coordination\n\nFiling as a private GHSA at `fedora-python/lxml_html_clean` \u2014 `lxml_html_clean` is the canonical maintainer of the `Cleaner` code (881 lines) and the security-responsible team since the lxml 5.2.0 split, where the cleaner was extracted out of lxml precisely so cleaner-security reports could land on the right team. The lxml side cannot be filed via GHSA (`https://github.com/lxml/lxml/security/advisories/new` returns 404 \u2014 private reporting is not enabled), so a parallel report has been emailed directly to the lxml maintainer for the upstream `defs.link_attrs` patch path. You\u0027re welcome to coordinate with them directly if you\u0027d prefer the upstream fix over the in-package alternative above.\n\nHappy to provide a draft patch or PR on either path. No bounty expected.",
  "id": "GHSA-4jhm-jv67-739f",
  "modified": "2026-07-08T20:23:22Z",
  "published": "2026-07-08T20:23:22Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/fedora-python/lxml_html_clean/security/advisories/GHSA-4jhm-jv67-739f"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/fedora-python/lxml_html_clean"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:L/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "`lxml_html_clean.Cleaner` does not strip `javascript:` URLs from namespaced URL attributes"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.





Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

Sightings

Author Source Type Date Other

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or observed by the user.
  • Confirmed: The vulnerability has been validated from an analyst's perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
  • Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
  • Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
  • Not confirmed: The user expressed doubt about the validity of the vulnerability.
  • Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.


Detection rules are retrieved from Rulezet.


Related by attack behaviour

Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.