GHSA-269C-H76Q-8CXW
Vulnerability from github – Published: 2026-09-17 17:28 – Updated: 2026-09-17 17:28Summary
A page editor without admin.super can place an event handler after a > inside a quoted attribute. Grav accepts and stores the page, then executes the handler in the application origin when a visitor opens it.
Details
Security::detectXss() (system/src/Grav/Common/Security.php:253) anchors the on_events scan at < and uses [^>]*?, which cannot cross the first literal >. When that character is inside a quoted value, the browser keeps the tag open and parses the later onerror attribute, so the detector and browser disagree. AdminController::savePage() relies on this detector when saving content from page editors outside the admin.super whitelist.
PoC
I reproduced this with getgrav/grav 2.0.11 (ad9709f865b09b68798fb1ac375b484a8cc1d892), Admin 1.10.52, and Quark 2 1.1.4.
- Sign in as a user with
admin.loginandadmin.pages, but withoutadmin.super. - Create or edit
/xsstestand save this page body:
<img src=x title=">" onerror=alert(document.domain)>
- Open
/xsstestin a private browser window.
The save succeeds and the visitor sees an alert containing the site domain. With the body changed to <img src=x onerror=alert(1)>, the same endpoint rejects it with XSS issue detected and does not store it.
Impact
A page editor can execute JavaScript in the origin of every user who views the stored page, including unauthenticated visitors.
Anticipated objection and response
Although the detectXss() docblock describes it as a heuristic that cannot catch every XSS, this check is the storage-time boundary for page editors outside the default security.xss_whitelist of admin.super. The same endpoint rejects a plain handler but accepts this executable form, allowing a lower-trust editor to cross the boundary the check is intended to enforce.
Suggested fix
Prefer an HTML tokenizer or sanitizer that rejects event-handler attributes on parsed elements. If the existing tripwire remains, make its tag scan quote-aware instead of treating every > as a boundary. Add double-quoted and single-quoted regression cases plus the rejected plain-handler control.
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 2.0.12"
},
"package": {
"ecosystem": "Packagist",
"name": "getgrav/grav"
},
"ranges": [
{
"events": [
{
"introduced": "1.5.2"
},
{
"fixed": "2.0.13"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-72832"
],
"database_specific": {
"cwe_ids": [
"CWE-79"
],
"github_reviewed": true,
"github_reviewed_at": "2026-09-17T17:28:34Z",
"nvd_published_at": null,
"severity": "MODERATE"
},
"details": "### Summary\n\nA page editor without `admin.super` can place an event handler after a `\u003e` inside a quoted attribute. Grav accepts and stores the page, then executes the handler in the application origin when a visitor opens it.\n\n### Details\n\n`Security::detectXss()` (`system/src/Grav/Common/Security.php:253`) anchors the `on_events` scan at `\u003c` and uses `[^\u003e]*?`, which cannot cross the first literal `\u003e`. When that character is inside a quoted value, the browser keeps the tag open and parses the later `onerror` attribute, so the detector and browser disagree. `AdminController::savePage()` relies on this detector when saving content from page editors outside the `admin.super` whitelist.\n\n### PoC\n\nI reproduced this with `getgrav/grav` 2.0.11 (`ad9709f865b09b68798fb1ac375b484a8cc1d892`), Admin 1.10.52, and Quark 2 1.1.4.\n\n1. Sign in as a user with `admin.login` and `admin.pages`, but without `admin.super`.\n2. Create or edit `/xsstest` and save this page body:\n\n```html\n\u003cimg src=x title=\"\u003e\" onerror=alert(document.domain)\u003e\n```\n\n3. Open `/xsstest` in a private browser window.\n\nThe save succeeds and the visitor sees an alert containing the site domain. With the body changed to `\u003cimg src=x onerror=alert(1)\u003e`, the same endpoint rejects it with `XSS issue detected` and does not store it.\n\n### Impact\n\nA page editor can execute JavaScript in the origin of every user who views the stored page, including unauthenticated visitors.\n\n### Anticipated objection and response\n\nAlthough the `detectXss()` docblock describes it as a heuristic that cannot catch every XSS, this check is the storage-time boundary for page editors outside the default `security.xss_whitelist` of `admin.super`. The same endpoint rejects a plain handler but accepts this executable form, allowing a lower-trust editor to cross the boundary the check is intended to enforce.\n\n### Suggested fix\n\nPrefer an HTML tokenizer or sanitizer that rejects event-handler attributes on parsed elements. If the existing tripwire remains, make its tag scan quote-aware instead of treating every `\u003e` as a boundary. Add double-quoted and single-quoted regression cases plus the rejected plain-handler control.",
"id": "GHSA-269c-h76q-8cxw",
"modified": "2026-09-17T17:28:34Z",
"published": "2026-09-17T17:28:34Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/getgrav/grav/security/advisories/GHSA-269c-h76q-8cxw"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-72832"
},
{
"type": "WEB",
"url": "https://github.com/getgrav/grav/commit/ad9709f865b09b68798fb1ac375b484a8cc1d892"
},
{
"type": "PACKAGE",
"url": "https://github.com/getgrav/grav"
},
{
"type": "WEB",
"url": "https://www.vulncheck.com/advisories/grav-before-stored-xss-via-quoted-attribute-bypass"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N",
"type": "CVSS_V3"
}
],
"summary": "Grav: Stored XSS via quoted-attribute bypass in detectXss"
}
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.