{"id":"CVE-2026-83616","aliases":["AZL-99351","CLSA-2026-1788980919","CLSA-2026-1790108624","CLSA-2026-1790755546","CLSA-2026-1790760268","CLSA-2026-1791987856","DEBIAN-CVE-2026-83616","EUVD-2026-69239","GHSA-c7q8-3ch8-vqpv","UBUNTU-CVE-2026-83616"],"status":"published","title":"xmldom: Processing Instruction Target Injection Bypasses requireWellFormed","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","severity":{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N","level":"high","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-12","score":0.00612,"percentile":0.49808,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-91"],"published":"2026-09-01T14:45:44Z","modified":"2026-11-05T13:28:23Z","fixes":[{"ecosystem":"npm","package":"@xmldom/xmldom","fixed_in":["0.8.15","0.9.12"],"sources":["ghsa","osv"]},{"ecosystem":"Debian:14","package":"node-xmldom","fixed_in":["0.9.12-1"],"sources":["osv"]}],"affected":[{"source":"ghsa","ecosystem":"npm","package":"@xmldom/xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.7.0"},{"last_affected":"0.8.14"},{"fixed":"0.8.15"}]}]},{"source":"ghsa","ecosystem":"npm","package":"@xmldom/xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.9.0"},{"last_affected":"0.9.11"},{"fixed":"0.9.12"}]}]},{"source":"ghsa","ecosystem":"npm","package":"xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"0.6.0"}]}]},{"source":"osv","ecosystem":"Azure Linux:3","package":"python-tensorboard","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"2.16.2-6"}]}]},{"source":"osv","ecosystem":"TuxCare:npm","package":"xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.6.0-tuxcare.0"},{"fixed":"0.6.0-tuxcare.4"}]}],"versions":["0.6.0","0.6.0-tuxcare.1","0.6.0-tuxcare.2","0.6.0-tuxcare.3"],"bundled":5},{"source":"osv","ecosystem":"TuxCare:npm","package":"xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.1.31-tuxcare.0"},{"fixed":"0.1.31-tuxcare.5"}]}],"versions":["0.1.31","0.1.31-tuxcare.1","0.1.31-tuxcare.2","0.1.31-tuxcare.3","0.1.31-tuxcare.4"],"bundled":5},{"source":"osv","ecosystem":"TuxCare:npm","package":"@xmldom/xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.7.13-tuxcare.0"},{"fixed":"0.7.13-tuxcare.1"}]}],"versions":["0.7.13"],"bundled":14},{"source":"osv","ecosystem":"TuxCare:npm","package":"@xmldom/xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.9.10-tuxcare.0"},{"fixed":"0.9.10-tuxcare.2"}]}],"versions":["0.9.10","0.9.10-tuxcare.1"],"bundled":4},{"source":"osv","ecosystem":"TuxCare:npm","package":"@xmldom/xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.9.8-tuxcare.0"},{"fixed":"0.9.8-tuxcare.5"}]}],"versions":["0.9.8","0.9.8-tuxcare.1","0.9.8-tuxcare.2","0.9.8-tuxcare.3","0.9.8-tuxcare.4"],"bundled":4},{"source":"osv","ecosystem":"Debian:12","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"}]}],"versions":["0.8.6-1","0.9.10-1","0.9.10-2","0.9.11-1","0.9.12-1","0.9.5-1","0.9.6-1","0.9.8-1","0.9.8-2","0.9.9-1"]},{"source":"osv","ecosystem":"Debian:13","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"}]}],"versions":["0.9.10-1","0.9.10-2","0.9.11-1","0.9.12-1","0.9.6-1","0.9.8-1","0.9.8-2","0.9.9-1"]},{"source":"osv","ecosystem":"Debian:14","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.9.12-1"}]}],"versions":["0.9.10-1","0.9.10-2","0.9.11-1","0.9.6-1","0.9.8-1","0.9.8-2","0.9.9-1"]},{"source":"osv","ecosystem":"npm","package":"@xmldom/xmldom","ranges":[{"type":"SEMVER","events":[{"introduced":"0.7.0"},{"fixed":"0.8.15"}]}]},{"source":"osv","ecosystem":"npm","package":"@xmldom/xmldom","ranges":[{"type":"SEMVER","events":[{"introduced":"0.9.0"},{"fixed":"0.9.12"}]}]},{"source":"osv","ecosystem":"npm","package":"xmldom","ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"last_affected":"0.6.0"}]}]},{"source":"osv","ecosystem":"Ubuntu:20.04:LTS","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"}]}],"versions":["0.1.27+ds-1","0.1.27+ds-1+deb10u2build0.20.04.1"]},{"source":"osv","ecosystem":"Ubuntu:22.04:LTS","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"}]}],"versions":["0.5.0-1","0.7.3-1","0.7.5-1","0.7.5-1ubuntu0.22.04.1"]},{"source":"osv","ecosystem":"Ubuntu:24.04:LTS","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"}]}],"versions":["0.8.6-1"]},{"source":"osv","ecosystem":"Ubuntu:26.10","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"}]}],"versions":["0.9.8-2","0.9.10-2","0.9.11-1"]},{"source":"osv","ecosystem":"Ubuntu:26.04:LTS","package":"node-xmldom","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"}]}],"versions":["0.9.6-1","0.9.8-1","0.9.8-2"]}],"products":[{"vendor":"@xmldom","product":"xmldom"},{"vendor":"xmldom","product":"xmldom"}],"references":[{"url":"https://github.com/xmldom/xmldom/security/advisories/GHSA-c7q8-3ch8-vqpv","tags":["x_refsource_CONFIRM","exploit","advisory","web","report"]},{"url":"https://github.com/xmldom/xmldom/pull/1071","tags":["x_refsource_MISC","fix","web","report"]},{"url":"https://github.com/xmldom/xmldom/pull/1072","tags":["x_refsource_MISC","fix","web","report"]},{"url":"https://github.com/xmldom/xmldom/commit/1cde3e31a07c41c87cfd368d6946aa477f16b4f9","tags":["x_refsource_MISC","fix","web"]},{"url":"https://github.com/xmldom/xmldom/commit/3b694872bcb5c7e3cbadba961a4be2488750ce5b","tags":["x_refsource_MISC","fix","web"]},{"url":"https://github.com/xmldom/xmldom/releases/tag/0.8.15","tags":["x_refsource_MISC","web"]},{"url":"https://github.com/xmldom/xmldom/releases/tag/0.9.12","tags":["x_refsource_MISC","web"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-83616","tags":["web","advisory"]},{"url":"https://github.com/advisories/GHSA-c7q8-3ch8-vqpv"},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/83xxx/CVE-2026-83616.json","tags":["advisory"]},{"url":"https://security-tracker.debian.org/tracker/CVE-2026-83616","tags":["advisory"]},{"url":"https://github.com/xmldom/xmldom","tags":["package"]},{"url":"https://ubuntu.com/security/CVE-2026-83616","tags":["report"]},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-83616","tags":["report"]}],"sources":{"cve":[{"id":"CVE-2026-83616","title":"xmldom: Processing Instruction Target Injection Bypasses requireWellFormed","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","status":"published","published":"2026-09-01T14:45:44.504Z","modified":"2026-09-01T15:18:45.24Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N","level":"high","assessor":"cna"}],"cwes":["CWE-91"],"updated_at":"2026-11-02T13:44:51.734189Z"}],"euvd":[{"id":"EUVD-2026-69239","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","status":"published","published":"2026-09-01T14:45:44Z","modified":"2026-09-01T15:18:45Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N","level":"high"}],"updated_at":"2026-11-02T18:01:17.79788Z"}],"ghsa":[{"id":"GHSA-c7q8-3ch8-vqpv","title":"xmldom: Processing Instruction Target Injection Bypasses requireWellFormed","description":"## Summary\n\n`Document.createProcessingInstruction()` in `@xmldom/xmldom` performs no validation on the `target` parameter. The `requireWellFormed: true` serializer option validates only for `:` in the target and a case-insensitive `xml` prefix, but does not check for `\u003e` characters. A `\u003e` in the target breaks the processing instruction boundary (`\u003c?...?\u003e`), allowing injection of arbitrary content into the serialized XML output.\n\n## Details\n\n`Document.createProcessingInstruction(target, data)` at `lib/dom.js` around line 2413 accepts any string as the `target` parameter and stores it on the PI node without validation.\n\nDuring serialization, the `requireWellFormed` code path (around line 3286) performs two checks on PI targets:\n\n1. Rejects targets containing `:` (namespace prefix check)\n2. Rejects targets matching `xml` case-insensitively (reserved prefix)\n\nHowever, it does NOT validate that the target conforms to the XML Name production, and critically does NOT check for `\u003e` characters. Since processing instructions are serialized as `\u003c?target data?\u003e`, a `\u003e` in the target prematurely closes the PI, causing the remaining content to be interpreted as document content by any downstream XML parser.\n\n### Root Cause\n\n1. `createProcessingInstruction()` performs no validation on `target`\n2. The serializer's `requireWellFormed` check is incomplete -- it only checks for `:` and `xml`, missing characters that break PI syntax (`\u003e`, `?`, whitespace)\n3. The serializer emits the target verbatim: `\u003c?${target} ${data}?\u003e`\n\n## Proof of Concept\n\n```javascript\nconst { DOMImplementation, XMLSerializer } = require('@xmldom/xmldom');\n\nconst impl = new DOMImplementation();\nconst serializer = new XMLSerializer();\nconst doc = impl.createDocument(null, 'root', null);\n\n// PI target containing \u003e breaks the PI boundary\nconst pi = doc.createProcessingInstruction('a\u003e', 'data');\ndoc.documentElement.appendChild(pi);\n\nconst output = serializer.serializeToString(doc, { requireWellFormed: true });\nconsole.log(output);\n// Output: \u003croot\u003e\u003c?a\u003e data?\u003e\u003c/root\u003e\n//\n// The \u003e in the target closes the PI prematurely.\n// A downstream XML parser sees:\n//   - Processing instruction: \u003c?a?\u003e  (target \"a\", no data)\n//   - Text content: \" data?\u003e\"\n//\n// requireWellFormed: true did NOT prevent the injection.\n```\n\n### Injecting elements via PI target\n\n```javascript\nconst pi2 = doc.createProcessingInstruction(\n  'a?\u003e\u003cscript xmlns=\"http://www.w3.org/1999/xhtml\"\u003ealert(1)\u003c/script\u003e\u003c?b',\n  ''\n);\ndoc.documentElement.appendChild(pi2);\n\nconst output2 = serializer.serializeToString(doc, { requireWellFormed: true });\nconsole.log(output2);\n// Output includes:\n//   \u003c?a?\u003e\u003cscript xmlns=\"http://www.w3.org/1999/xhtml\"\u003ealert(1)\u003c/script\u003e\u003c?b ?\u003e\n//\n// The injected \u003cscript\u003e element is valid XHTML that a browser would execute.\n```\n\n## Impact\n\nApplications that create processing instructions with user-controlled target strings and serialize the result are vulnerable to XML injection. This enables:\n\n- **XML structure injection**: Breaking the PI boundary to inject arbitrary elements, text, or additional processing instructions into the output\n- **XSS via XHTML**: If the serialized output is served as XHTML or processed by a browser-based XML parser, injected script elements will execute\n- **XXE chain**: Injected DOCTYPE declarations or entity references could trigger XXE in downstream XML parsers that consume the output\n- **requireWellFormed bypass**: The existing well-formedness checks are incomplete and provide a false sense of security\n\n## Fix Applied\n\nUnder `requireWellFormed`, the serializer validates a processing-instruction target as an XML `NCName` (a `Name` with no colon) and rejects a case-insensitive `xml`, throwing `InvalidStateError` when the target is ill-formed — so a `\u003e`, `?`, or whitespace in the target is now refused.\\\nOn 0.9.12 this replaces an earlier check that already rejected a colon or `xml`, so the no-colon rule is preserved.\\\n0.8.15 had no processing-instruction target check at all, so the whole target validation is new there.\\\nNon-breaking and opt-in. See the [XML `Name` production](https://www.w3.org/TR/xml/#NT-Name).\n\u003e **⚠ 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```javascript\nconst { DOMImplementation, XMLSerializer } = require('@xmldom/xmldom');\n\nconst impl = new DOMImplementation();\nconst serializer = new XMLSerializer();\nconst doc = impl.createDocument(null, 'root', null);\n\n// PI target containing \u003e breaks the PI boundary\nconst pi = doc.createProcessingInstruction('a\u003e', 'data');\ndoc.documentElement.appendChild(pi);\n\n// Default path: emits the ill-formed target verbatim.\nconsole.log(serializer.serializeToString(doc));\n// Output: \u003croot\u003e\u003c?a\u003e data?\u003e\u003c/root\u003e\n\n// Opt-in path: the target check now rejects the break-out character.\ntry {\n  serializer.serializeToString(doc, { requireWellFormed: true });\n} catch (e) {\n  console.log(e.name); // InvalidStateError\n}\n```\n\n### Why the default stays verbatim\n\nW3C DOM Parsing's require-well-formed flag defaults to false, and the browser `XMLSerializer` emits the target verbatim in that default mode. Unconditionally throwing on an ill-formed PI target would diverge from that platform behavior and would be an unjustified breaking change, so the stricter validation is gated behind `{ requireWellFormed: true }`. (See the [W3C XML Name production](https://www.w3.org/TR/xml/#NT-Name) and [XML Processing Instructions](https://www.w3.org/TR/xml/#sec-pi).)\n\n### Residual limitation\n\nThe default serialization path still emits the ill-formed target verbatim -- only the opt-in `requireWellFormed` path is protected. Creation-time validation of the `target` in `createProcessingInstruction()` is breaking and is deferred to the next breaking release, tracked at [xmldom/xmldom#1073](https://github.com/xmldom/xmldom/issues/1073).","status":"published","published":"2026-09-08T21:03:28Z","modified":"2026-09-08T21:03:31Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N","level":"high"}],"cwes":["CWE-91"],"updated_at":"2026-11-02T14:41:14.764963Z"}],"nvd":[{"id":"CVE-2026-83616","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","status":"published","published":"2026-09-01T15:17:40.04Z","modified":"2026-09-08T21:11:31.703Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X","level":"high","assessor":"secondary"}],"cwes":["CWE-91"],"updated_at":"2026-11-02T14:12:43.04945Z"}],"osv":[{"id":"CVE-2026-83616","title":"xmldom: Processing Instruction Target Injection Bypasses requireWellFormed","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","status":"published","published":"2026-09-01T14:45:44.504Z","modified":"2026-11-11T02:49:03.934378218Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N","level":"high"}],"cwes":["CWE-91"],"updated_at":"2026-11-11T03:21:28.618498Z"},{"id":"DEBIAN-CVE-2026-83616","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","status":"published","published":"2026-09-01T15:17:40.04Z","modified":"2026-10-01T03:03:43.210463569Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X","level":"high"}],"updated_at":"2026-11-02T15:46:04.577397Z"},{"id":"GHSA-c7q8-3ch8-vqpv","title":"xmldom: Processing Instruction Target Injection Bypasses requireWellFormed","description":"## Summary\n\n`Document.createProcessingInstruction()` in `@xmldom/xmldom` performs no validation on the `target` parameter. The `requireWellFormed: true` serializer option validates only for `:` in the target and a case-insensitive `xml` prefix, but does not check for `\u003e` characters. A `\u003e` in the target breaks the processing instruction boundary (`\u003c?...?\u003e`), allowing injection of arbitrary content into the serialized XML output.\n\n## Details\n\n`Document.createProcessingInstruction(target, data)` at `lib/dom.js` around line 2413 accepts any string as the `target` parameter and stores it on the PI node without validation.\n\nDuring serialization, the `requireWellFormed` code path (around line 3286) performs two checks on PI targets:\n\n1. Rejects targets containing `:` (namespace prefix check)\n2. Rejects targets matching `xml` case-insensitively (reserved prefix)\n\nHowever, it does NOT validate that the target conforms to the XML Name production, and critically does NOT check for `\u003e` characters. Since processing instructions are serialized as `\u003c?target data?\u003e`, a `\u003e` in the target prematurely closes the PI, causing the remaining content to be interpreted as document content by any downstream XML parser.\n\n### Root Cause\n\n1. `createProcessingInstruction()` performs no validation on `target`\n2. The serializer's `requireWellFormed` check is incomplete -- it only checks for `:` and `xml`, missing characters that break PI syntax (`\u003e`, `?`, whitespace)\n3. The serializer emits the target verbatim: `\u003c?${target} ${data}?\u003e`\n\n## Proof of Concept\n\n```javascript\nconst { DOMImplementation, XMLSerializer } = require('@xmldom/xmldom');\n\nconst impl = new DOMImplementation();\nconst serializer = new XMLSerializer();\nconst doc = impl.createDocument(null, 'root', null);\n\n// PI target containing \u003e breaks the PI boundary\nconst pi = doc.createProcessingInstruction('a\u003e', 'data');\ndoc.documentElement.appendChild(pi);\n\nconst output = serializer.serializeToString(doc, { requireWellFormed: true });\nconsole.log(output);\n// Output: \u003croot\u003e\u003c?a\u003e data?\u003e\u003c/root\u003e\n//\n// The \u003e in the target closes the PI prematurely.\n// A downstream XML parser sees:\n//   - Processing instruction: \u003c?a?\u003e  (target \"a\", no data)\n//   - Text content: \" data?\u003e\"\n//\n// requireWellFormed: true did NOT prevent the injection.\n```\n\n### Injecting elements via PI target\n\n```javascript\nconst pi2 = doc.createProcessingInstruction(\n  'a?\u003e\u003cscript xmlns=\"http://www.w3.org/1999/xhtml\"\u003ealert(1)\u003c/script\u003e\u003c?b',\n  ''\n);\ndoc.documentElement.appendChild(pi2);\n\nconst output2 = serializer.serializeToString(doc, { requireWellFormed: true });\nconsole.log(output2);\n// Output includes:\n//   \u003c?a?\u003e\u003cscript xmlns=\"http://www.w3.org/1999/xhtml\"\u003ealert(1)\u003c/script\u003e\u003c?b ?\u003e\n//\n// The injected \u003cscript\u003e element is valid XHTML that a browser would execute.\n```\n\n## Impact\n\nApplications that create processing instructions with user-controlled target strings and serialize the result are vulnerable to XML injection. This enables:\n\n- **XML structure injection**: Breaking the PI boundary to inject arbitrary elements, text, or additional processing instructions into the output\n- **XSS via XHTML**: If the serialized output is served as XHTML or processed by a browser-based XML parser, injected script elements will execute\n- **XXE chain**: Injected DOCTYPE declarations or entity references could trigger XXE in downstream XML parsers that consume the output\n- **requireWellFormed bypass**: The existing well-formedness checks are incomplete and provide a false sense of security\n\n## Fix Applied\n\nUnder `requireWellFormed`, the serializer validates a processing-instruction target as an XML `NCName` (a `Name` with no colon) and rejects a case-insensitive `xml`, throwing `InvalidStateError` when the target is ill-formed — so a `\u003e`, `?`, or whitespace in the target is now refused.\\\nOn 0.9.12 this replaces an earlier check that already rejected a colon or `xml`, so the no-colon rule is preserved.\\\n0.8.15 had no processing-instruction target check at all, so the whole target validation is new there.\\\nNon-breaking and opt-in. See the [XML `Name` production](https://www.w3.org/TR/xml/#NT-Name).\n\u003e **⚠ 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```javascript\nconst { DOMImplementation, XMLSerializer } = require('@xmldom/xmldom');\n\nconst impl = new DOMImplementation();\nconst serializer = new XMLSerializer();\nconst doc = impl.createDocument(null, 'root', null);\n\n// PI target containing \u003e breaks the PI boundary\nconst pi = doc.createProcessingInstruction('a\u003e', 'data');\ndoc.documentElement.appendChild(pi);\n\n// Default path: emits the ill-formed target verbatim.\nconsole.log(serializer.serializeToString(doc));\n// Output: \u003croot\u003e\u003c?a\u003e data?\u003e\u003c/root\u003e\n\n// Opt-in path: the target check now rejects the break-out character.\ntry {\n  serializer.serializeToString(doc, { requireWellFormed: true });\n} catch (e) {\n  console.log(e.name); // InvalidStateError\n}\n```\n\n### Why the default stays verbatim\n\nW3C DOM Parsing's require-well-formed flag defaults to false, and the browser `XMLSerializer` emits the target verbatim in that default mode. Unconditionally throwing on an ill-formed PI target would diverge from that platform behavior and would be an unjustified breaking change, so the stricter validation is gated behind `{ requireWellFormed: true }`. (See the [W3C XML Name production](https://www.w3.org/TR/xml/#NT-Name) and [XML Processing Instructions](https://www.w3.org/TR/xml/#sec-pi).)\n\n### Residual limitation\n\nThe default serialization path still emits the ill-formed target verbatim -- only the opt-in `requireWellFormed` path is protected. Creation-time validation of the `target` in `createProcessingInstruction()` is breaking and is deferred to the next breaking release, tracked at [xmldom/xmldom#1073](https://github.com/xmldom/xmldom/issues/1073).","status":"published","published":"2026-09-08T21:03:28Z","modified":"2026-09-08T21:15:04.899638684Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N","level":"high"}],"cwes":["CWE-91"],"updated_at":"2026-11-02T15:46:44.571161Z"},{"id":"UBUNTU-CVE-2026-83616","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","status":"published","published":"2026-09-01T15:17:00Z","modified":"2026-10-17T13:01:11.914735944Z","cvss":[{"version":"4.0","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N","level":"high"}],"updated_at":"2026-11-02T16:57:22.213888Z"},{"id":"AZL-99351","title":"CVE-2026-83616 affecting package python-tensorboard 2.16.2-6","description":"xmldom is a pure JavaScript W3C standard-based (XML DOM Level 2 Core) DOMParser and XMLSerializer module. Prior to @xmldom/xmldom versions 0.8.15 and 0.9.12, and in xmldom version 0.6.0 and earlier, Document.createProcessingInstruction(target, data) in lib/dom.js accepts an unvalidated target, while the requireWellFormed: true serializer checks only for a colon and the reserved case-insensitive xml name on 0.9.x and performs no target check on 0.8.x. Because serialization emits \u003c?target data?\u003e, a target containing \u003e, ?, whitespace, or another invalid XML-name character can break the processing-instruction boundary and inject XML structure. This issue is fixed in @xmldom/xmldom versions 0.8.15 and 0.9.12; no fixed version is available for xmldom.","status":"published","published":"2026-09-01T15:17:40Z","modified":"2026-09-08T05:27:28Z","updated_at":"2026-11-02T15:28:12.596039Z"},{"id":"CLSA-2026-1788980919","title":"TuxCare security update for xmldom (5 CVEs)","description":"TuxCare rebuilt xmldom to address 5 CVEs. Fixed in 0.6.0-tuxcare.4.","status":"published","published":"2026-09-09T19:08:39Z","modified":"2026-10-15T09:47:56.142670074Z","updated_at":"2026-11-02T15:43:47.938828Z"},{"id":"CLSA-2026-1790108624","title":"TuxCare security update for xmldom (5 CVEs)","description":"TuxCare rebuilt xmldom to address 5 CVEs. Fixed in 0.1.31-tuxcare.5.","status":"published","published":"2026-09-22T20:23:44Z","modified":"2026-11-05T13:28:23.742251042Z","updated_at":"2026-11-05T16:37:45.64821Z"},{"id":"CLSA-2026-1790755546","title":"TuxCare security update for @xmldom/xmldom (14 CVEs)","description":"TuxCare rebuilt @xmldom/xmldom to address 14 CVEs. Fixed in 0.7.13-tuxcare.1.","status":"published","published":"2026-09-30T08:05:46Z","modified":"2026-10-15T09:47:53.549832566Z","updated_at":"2026-11-02T15:43:48.65893Z"},{"id":"CLSA-2026-1790760268","title":"TuxCare security update for @xmldom/xmldom (4 CVEs)","description":"TuxCare rebuilt @xmldom/xmldom to address 4 CVEs. Fixed in 0.9.10-tuxcare.2.","status":"published","published":"2026-09-30T09:24:28Z","modified":"2026-10-15T09:47:57.301956488Z","updated_at":"2026-11-02T15:43:48.65893Z"},{"id":"CLSA-2026-1791987856","title":"TuxCare security update for @xmldom/xmldom (4 CVEs)","description":"TuxCare rebuilt @xmldom/xmldom to address 4 CVEs. Fixed in 0.9.8-tuxcare.5.","status":"published","published":"2026-10-14T14:24:16Z","modified":"2026-10-15T09:48:00.183619915Z","updated_at":"2026-11-02T15:43:50.599064Z"}]}}