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

GHSA-6GMQ-8VP8-GCM6

Vulnerability from github – Published: 2026-09-02 15:18 – Updated: 2026-09-02 15:18
VLAI
Summary
xmldom: XML fragment injection via invalid EntityReference.nodeName during requireWellFormed serialization
Details

Summary

An EntityReference node can be created with an invalid, attacker-controlled name through Document.createEntityReference(name). When this node is serialized directly with:

serializer.serializeToString(ref, { requireWellFormed: true })

the invalid nodeName is emitted into the serialized XML fragment without validation or escaping.

This can produce real XML markup in the serialized output. In the proof of concept below, the serialized fragment contains <injected/>, and reparsing the fragment creates a real injected element.


Details

The issue appears to be in the serialization path for ENTITY_REFERENCE_NODE.

For several other node types, requireWellFormed: true performs specific validation checks before serialization. For example, comments, processing instructions, document types, and some character data cases are checked before being emitted.

However, for ENTITY_REFERENCE_NODE, the serializer appears to emit the node name directly in entity reference form:

case ENTITY_REFERENCE_NODE:
  buf.push('&', n.nodeName, ';');
  return null;

As a result, if nodeName contains characters that break out of the intended &name; structure, the serializer can emit additional XML markup.

For example, an entity reference created with the name:

safe; <injected/> &x

is serialized as:

&safe; <injected/> &x;

When this fragment is later parsed in an XML context, <injected/> becomes a real element.

This is especially surprising when { requireWellFormed: true } is used, because applications may reasonably treat this mode as the stricter or safer XML serialization mode.


Proof of Concept

Tested with:

@xmldom/xmldom@0.9.10
Node.js v24.18.0
Windows 10 / PowerShell
'use strict';

const { DOMImplementation, XMLSerializer, DOMParser } = require('@xmldom/xmldom');

const impl = new DOMImplementation();
const doc = impl.createDocument(null, 'root', null);
const serializer = new XMLSerializer();

function countInjected(fragment) {
  try {
    const parsed = new DOMParser().parseFromString(`<root>${fragment}</root>`, 'application/xml');
    return parsed.getElementsByTagName('injected').length;
  } catch (e) {
    return `PARSE_THROW ${e.name}: ${e.message}`;
  }
}

for (const name of [
  'safe',
  'safe; <injected/> &x',
  'x<injected',
  'x y'
]) {
  try {
    const ref = doc.createEntityReference(name);
    const xml = serializer.serializeToString(ref, { requireWellFormed: true });

    console.log(`[SERIALIZED] ${JSON.stringify(name)}: ${xml}`);
    console.log(`[INJECTED_COUNT] ${JSON.stringify(name)}: ${countInjected(xml)}`);
  } catch (e) {
    console.log(`[THROW] ${JSON.stringify(name)}: ${e.name}: ${e.message}`);
  }
}

Observed output:

[SERIALIZED] "safe": &safe;
[INJECTED_COUNT] "safe": 0

[SERIALIZED] "safe; <injected/> &x": &safe; <injected/> &x;
[INJECTED_COUNT] "safe; <injected/> &x": 1

[SERIALIZED] "x<injected": &x<injected;
[INJECTED_COUNT] "x<injected": 0

[SERIALIZED] "x y": &x y;
[INJECTED_COUNT] "x y": 0

Impact

An application that creates an EntityReference from attacker-controlled input and then serializes that node or XML fragment with requireWellFormed: true may produce XML containing attacker-controlled markup.

The impact is limited by two observations:

  1. The parser does not create EntityReference nodes from ordinary XML entity references.
  2. Appending an EntityReference node as an element child is rejected with a HierarchyRequestError.

The main affected scenario is applications that directly use createEntityReference(name) and then serialize the resulting node or fragment.

Fix Applied

Two complementary, non-breaking fixes. (1) document.createEntityReference(name) rejects an invalid Name at creation, closing the reachable creation vector by default — the opt-in serializer check alone cannot, since a later nodeName mutation would bypass a creation-only guard. (2) Under requireWellFormed, the serializer validates the EntityReference nodeName as a well-formed XML Name and throws InvalidStateError when it is not; a valid reference still serializes as &name;. Both ship on both maintained versions. The EntityReference / createEntityReference docs note that under requireWellFormed the nodeName is validated as an XML Name, and that xmldom does not expand entities. See the XML Name production.

⚠ Opt-in required. Protection is not automatic. Existing serialization calls remain vulnerable unless { requireWellFormed: true } is explicitly passed. Applications that serialize untrusted DOM content should audit all serializeToString() call sites and add it.

Proof of Concept - fixed path

'use strict';

const { DOMImplementation, XMLSerializer } = require('@xmldom/xmldom');

const impl = new DOMImplementation();
const doc = impl.createDocument(null, 'root', null);
const serializer = new XMLSerializer();

// Creation-time anchor (applied by default): an invalid XML Name is rejected at creation.
try {
  doc.createEntityReference('safe; <injected/> &x');
} catch (e) {
  console.log(`${e.name}`); // rejected at creation
}

// Default path (requireWellFormed omitted): because creation now rejects an ill-formed name,
// an ill-formed nodeName is only reachable via a post-creation mutation — and is emitted verbatim.
const ref = doc.createEntityReference('safe');
ref.nodeName = 'safe; <injected/> &x';
console.log(serializer.serializeToString(ref));
// -> &safe; <injected/> &x;   (injection present on the default path)

// Opt-in path: throws on the invalid nodeName.
try {
  serializer.serializeToString(ref, { requireWellFormed: true });
} catch (e) {
  console.log(`${e.name}`); // InvalidStateError
}

// A valid name still serializes as &name; under requireWellFormed.
const ok = doc.createEntityReference('valid');
console.log(serializer.serializeToString(ok, { requireWellFormed: true }));
// -> &valid;

Why the default stays verbatim

The creation-time anchor is applied by default, because it is classified non-breaking. The serializer check, by contrast, stays gated behind { requireWellFormed: true }: W3C DOM Parsing's require-well-formed flag defaults to false, and the browser XMLSerializer emits the nodeName verbatim in that default mode, so unconditionally throwing for an ill-formed EntityReference.nodeName would be an unjustified breaking change — which is why the default serialization path stays verbatim.

Residual limitation

The creation vector is closed by default — the non-breaking creation-time anchor — with no further deferred work. The residual is at serialization: the default path still emits an ill-formed nodeName verbatim, because the serializer check is opt-in via { requireWellFormed: true }.

Show details on source website

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 0.8.14"
      },
      "package": {
        "ecosystem": "npm",
        "name": "@xmldom/xmldom"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0.7.0"
            },
            {
              "fixed": "0.8.15"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 0.9.11"
      },
      "package": {
        "ecosystem": "npm",
        "name": "@xmldom/xmldom"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0.9.0"
            },
            {
              "fixed": "0.9.12"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "npm",
        "name": "xmldom"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "last_affected": "0.6.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-83610"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-116"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-09-02T15:18:20Z",
    "nvd_published_at": "2026-09-01T15:17:39Z",
    "severity": "MODERATE"
  },
  "details": "## Summary\n\nAn `EntityReference` node can be created with an invalid, attacker-controlled name through `Document.createEntityReference(name)`. When this node is serialized directly with:\n\n```js\nserializer.serializeToString(ref, { requireWellFormed: true })\n```\n\nthe invalid `nodeName` is emitted into the serialized XML fragment without validation or escaping.\n\nThis can produce real XML markup in the serialized output. In the proof of concept below, the serialized fragment contains `\u003cinjected/\u003e`, and reparsing the fragment creates a real `injected` element.\n\n---\n\n## Details\n\nThe issue appears to be in the serialization path for `ENTITY_REFERENCE_NODE`.\n\nFor several other node types, `requireWellFormed: true` performs specific validation checks before serialization. For example, comments, processing instructions, document types, and some character data cases are checked before being emitted.\n\nHowever, for `ENTITY_REFERENCE_NODE`, the serializer appears to emit the node name directly in entity reference form:\n\n```js\ncase ENTITY_REFERENCE_NODE:\n  buf.push(\u0027\u0026\u0027, n.nodeName, \u0027;\u0027);\n  return null;\n```\n\nAs a result, if `nodeName` contains characters that break out of the intended `\u0026name;` structure, the serializer can emit additional XML markup.\n\nFor example, an entity reference created with the name:\n\n```text\nsafe; \u003cinjected/\u003e \u0026x\n```\n\nis serialized as:\n\n```xml\n\u0026safe; \u003cinjected/\u003e \u0026x;\n```\n\nWhen this fragment is later parsed in an XML context, `\u003cinjected/\u003e` becomes a real element.\n\nThis is especially surprising when `{ requireWellFormed: true }` is used, because applications may reasonably treat this mode as the stricter or safer XML serialization mode.\n\n---\n\n## Proof of Concept\n\nTested with:\n\n```text\n@xmldom/xmldom@0.9.10\nNode.js v24.18.0\nWindows 10 / PowerShell\n```\n\n```js\n\u0027use strict\u0027;\n\nconst { DOMImplementation, XMLSerializer, DOMParser } = require(\u0027@xmldom/xmldom\u0027);\n\nconst impl = new DOMImplementation();\nconst doc = impl.createDocument(null, \u0027root\u0027, null);\nconst serializer = new XMLSerializer();\n\nfunction countInjected(fragment) {\n  try {\n    const parsed = new DOMParser().parseFromString(`\u003croot\u003e${fragment}\u003c/root\u003e`, \u0027application/xml\u0027);\n    return parsed.getElementsByTagName(\u0027injected\u0027).length;\n  } catch (e) {\n    return `PARSE_THROW ${e.name}: ${e.message}`;\n  }\n}\n\nfor (const name of [\n  \u0027safe\u0027,\n  \u0027safe; \u003cinjected/\u003e \u0026x\u0027,\n  \u0027x\u003cinjected\u0027,\n  \u0027x y\u0027\n]) {\n  try {\n    const ref = doc.createEntityReference(name);\n    const xml = serializer.serializeToString(ref, { requireWellFormed: true });\n\n    console.log(`[SERIALIZED] ${JSON.stringify(name)}: ${xml}`);\n    console.log(`[INJECTED_COUNT] ${JSON.stringify(name)}: ${countInjected(xml)}`);\n  } catch (e) {\n    console.log(`[THROW] ${JSON.stringify(name)}: ${e.name}: ${e.message}`);\n  }\n}\n```\n\nObserved output:\n\n```text\n[SERIALIZED] \"safe\": \u0026safe;\n[INJECTED_COUNT] \"safe\": 0\n\n[SERIALIZED] \"safe; \u003cinjected/\u003e \u0026x\": \u0026safe; \u003cinjected/\u003e \u0026x;\n[INJECTED_COUNT] \"safe; \u003cinjected/\u003e \u0026x\": 1\n\n[SERIALIZED] \"x\u003cinjected\": \u0026x\u003cinjected;\n[INJECTED_COUNT] \"x\u003cinjected\": 0\n\n[SERIALIZED] \"x y\": \u0026x y;\n[INJECTED_COUNT] \"x y\": 0\n```\n\n---\n\n## Impact\n\nAn application that creates an `EntityReference` from attacker-controlled input and then serializes that node or XML fragment with `requireWellFormed: true` may produce XML containing attacker-controlled markup.\n\nThe impact is limited by two observations:\n\n1. The parser does not create `EntityReference` nodes from ordinary XML entity references.\n2. Appending an `EntityReference` node as an element child is rejected with a `HierarchyRequestError`.\n\nThe main affected scenario is applications that directly use `createEntityReference(name)` and then serialize the resulting node or fragment.\n\n## Fix Applied\n\nTwo complementary, non-breaking fixes.\n(1) `document.createEntityReference(name)` rejects an invalid `Name` at creation, closing the reachable creation vector by default \u2014 the opt-in serializer check alone cannot, since a later `nodeName` mutation would bypass a creation-only guard.\n(2) Under `requireWellFormed`, the serializer validates the `EntityReference` `nodeName` as a well-formed XML `Name` and throws `InvalidStateError` when it is not; a valid reference still serializes as `\u0026name;`. Both ship on both maintained versions. The `EntityReference` / `createEntityReference` docs note that under `requireWellFormed` the `nodeName` is validated as an XML `Name`, and that xmldom does not expand entities. See the [XML `Name` production](https://www.w3.org/TR/xml/#NT-Name).\n\u003e **\u26a0 Opt-in required.** Protection is not automatic. Existing serialization calls remain\n\u003e vulnerable unless `{ requireWellFormed: true }` is explicitly passed. Applications that\n\u003e serialize untrusted DOM content should audit all `serializeToString()` call sites and add it.\n\n### Proof of Concept - fixed path\n\n```js\n\u0027use strict\u0027;\n\nconst { DOMImplementation, XMLSerializer } = require(\u0027@xmldom/xmldom\u0027);\n\nconst impl = new DOMImplementation();\nconst doc = impl.createDocument(null, \u0027root\u0027, null);\nconst serializer = new XMLSerializer();\n\n// Creation-time anchor (applied by default): an invalid XML Name is rejected at creation.\ntry {\n  doc.createEntityReference(\u0027safe; \u003cinjected/\u003e \u0026x\u0027);\n} catch (e) {\n  console.log(`${e.name}`); // rejected at creation\n}\n\n// Default path (requireWellFormed omitted): because creation now rejects an ill-formed name,\n// an ill-formed nodeName is only reachable via a post-creation mutation \u2014 and is emitted verbatim.\nconst ref = doc.createEntityReference(\u0027safe\u0027);\nref.nodeName = \u0027safe; \u003cinjected/\u003e \u0026x\u0027;\nconsole.log(serializer.serializeToString(ref));\n// -\u003e \u0026safe; \u003cinjected/\u003e \u0026x;   (injection present on the default path)\n\n// Opt-in path: throws on the invalid nodeName.\ntry {\n  serializer.serializeToString(ref, { requireWellFormed: true });\n} catch (e) {\n  console.log(`${e.name}`); // InvalidStateError\n}\n\n// A valid name still serializes as \u0026name; under requireWellFormed.\nconst ok = doc.createEntityReference(\u0027valid\u0027);\nconsole.log(serializer.serializeToString(ok, { requireWellFormed: true }));\n// -\u003e \u0026valid;\n```\n\n### Why the default stays verbatim\n\nThe creation-time anchor is applied by default, because it is classified non-breaking. The serializer check, by contrast, stays gated behind `{ requireWellFormed: true }`: W3C DOM Parsing\u0027s require-well-formed flag defaults to `false`, and the browser `XMLSerializer` emits the `nodeName` verbatim in that default mode, so unconditionally throwing for an ill-formed `EntityReference.nodeName` would be an unjustified breaking change \u2014 which is why the default serialization path stays verbatim.\n\n### Residual limitation\n\nThe creation vector is closed by default \u2014 the non-breaking creation-time anchor \u2014 with no further deferred work. The residual is at serialization: the default path still emits an ill-formed `nodeName` verbatim, because the serializer check is opt-in via `{ requireWellFormed: true }`.",
  "id": "GHSA-6gmq-8vp8-gcm6",
  "modified": "2026-09-02T15:18:20Z",
  "published": "2026-09-02T15:18:20Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/xmldom/xmldom/security/advisories/GHSA-6gmq-8vp8-gcm6"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-83610"
    },
    {
      "type": "WEB",
      "url": "https://github.com/xmldom/xmldom/pull/1071"
    },
    {
      "type": "WEB",
      "url": "https://github.com/xmldom/xmldom/pull/1072"
    },
    {
      "type": "WEB",
      "url": "https://github.com/xmldom/xmldom/commit/4664386e4f4d99d17b416a151dbe8323e245284b"
    },
    {
      "type": "WEB",
      "url": "https://github.com/xmldom/xmldom/commit/6c3fb5ffeafe7901ec928ce9010988dd716c94a0"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/xmldom/xmldom"
    },
    {
      "type": "WEB",
      "url": "https://github.com/xmldom/xmldom/releases/tag/0.8.15"
    },
    {
      "type": "WEB",
      "url": "https://github.com/xmldom/xmldom/releases/tag/0.9.12"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N",
      "type": "CVSS_V4"
    }
  ],
  "summary": "xmldom: XML fragment injection via invalid EntityReference.nodeName during requireWellFormed serialization"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

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.

Loading…

Detection rules are retrieved from Rulezet.

Loading…

Loading…

Loading…