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.
6020 vulnerabilities reference this CWE, most recent first.
GHSA-8WG4-H2WP-63R5
Vulnerability from github – Published: 2025-04-04 18:31 – Updated: 2025-04-04 18:31A vulnerability was found in xujiangfei admintwo 1.0. It has been classified as critical. Affected is an unknown function of the file /resource/add. The manipulation of the argument description leads to server-side request forgery. It is possible to launch the attack remotely. The exploit has been disclosed to the public and may be used.
{
"affected": [],
"aliases": [
"CVE-2025-3254"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2025-04-04T16:15:40Z",
"severity": "MODERATE"
},
"details": "A vulnerability was found in xujiangfei admintwo 1.0. It has been classified as critical. Affected is an unknown function of the file /resource/add. The manipulation of the argument description leads to server-side request forgery. It is possible to launch the attack remotely. The exploit has been disclosed to the public and may be used.",
"id": "GHSA-8wg4-h2wp-63r5",
"modified": "2025-04-04T18:31:07Z",
"published": "2025-04-04T18:31:07Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-3254"
},
{
"type": "WEB",
"url": "https://github.com/caigo8/CVE-md/blob/main/admintwo/SSRF.md"
},
{
"type": "WEB",
"url": "https://vuldb.com/?ctiid.303324"
},
{
"type": "WEB",
"url": "https://vuldb.com/?id.303324"
},
{
"type": "WEB",
"url": "https://vuldb.com/?submit.548979"
}
],
"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:L",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:X/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-8WMW-PRW8-2GGM
Vulnerability from github – Published: 2026-04-17 15:31 – Updated: 2026-04-24 21:03Craftql v1.3.7 and before is vulnerable to Server-Side Request Forgery (SSRF) which allows an attacker to execute arbitrary code via the vendor/markhuot/craftql/src/Listeners/GetAssetsFieldSchema.php file.
{
"affected": [
{
"package": {
"ecosystem": "Packagist",
"name": "markhuot/craftql"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "1.3.7"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-31317"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2026-04-18T01:03:36Z",
"nvd_published_at": "2026-04-17T14:16:33Z",
"severity": "MODERATE"
},
"details": "Craftql v1.3.7 and before is vulnerable to Server-Side Request Forgery (SSRF) which allows an attacker to execute arbitrary code via the vendor/markhuot/craftql/src/Listeners/GetAssetsFieldSchema.php file.",
"id": "GHSA-8wmw-prw8-2ggm",
"modified": "2026-04-24T21:03:56Z",
"published": "2026-04-17T15:31:18Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-31317"
},
{
"type": "PACKAGE",
"url": "https://github.com/markhuot/craftql"
},
{
"type": "WEB",
"url": "https://github.com/stormmmg/craftql_ssrf"
},
{
"type": "WEB",
"url": "https://github.com/stormmmg/craftql_ssrf/blob/master/craftql-ssrf-en/README_detail.md"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:N/VA:N/SC:L/SI:L/SA:N",
"type": "CVSS_V4"
}
],
"summary": "Craftql vulnerable to Server-Side Request Forgery"
}
GHSA-8WX4-4FR5-G6R8
Vulnerability from github – Published: 2025-06-03 18:30 – Updated: 2025-06-03 18:30A vulnerability classified as critical was found in quequnlong shiyi-blog up to 1.2.1. This vulnerability affects unknown code of the file /app/sys/article/optimize. The manipulation of the argument url leads to server-side request forgery. The attack can be initiated remotely. The exploit has been disclosed to the public and may be used. The vendor was contacted early about this disclosure but did not respond in any way.
{
"affected": [],
"aliases": [
"CVE-2025-5510"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2025-06-03T17:15:22Z",
"severity": "MODERATE"
},
"details": "A vulnerability classified as critical was found in quequnlong shiyi-blog up to 1.2.1. This vulnerability affects unknown code of the file /app/sys/article/optimize. The manipulation of the argument url leads to server-side request forgery. The attack can be initiated remotely. The exploit has been disclosed to the public and may be used. The vendor was contacted early about this disclosure but did not respond in any way.",
"id": "GHSA-8wx4-4fr5-g6r8",
"modified": "2025-06-03T18:30:41Z",
"published": "2025-06-03T18:30:41Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-5510"
},
{
"type": "WEB",
"url": "https://github.com/uglory-gll/javasec/blob/main/shiyi-blog.md"
},
{
"type": "WEB",
"url": "https://github.com/uglory-gll/javasec/blob/main/shiyi-blog.md#2ssrf"
},
{
"type": "WEB",
"url": "https://vuldb.com/?ctiid.310924"
},
{
"type": "WEB",
"url": "https://vuldb.com/?id.310924"
},
{
"type": "WEB",
"url": "https://vuldb.com/?submit.584489"
}
],
"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:L",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:X/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-8X3Q-JPJH-QH5C
Vulnerability from github – Published: 2026-08-31 20:28 – Updated: 2026-08-31 20:28Summary
elFinder 2.1.69 is vulnerable to a Server-Side Request Forgery (SSRF) protection bypass when PHP cURL is unavailable and URL uploads use the fsock_get_contents() fallback.
An attacker who can submit a URL for server-side upload can use an attacker-controlled DNS hostname that initially resolves to an allowed public IP address and subsequently resolves to a loopback or private IP address.
The URL validation checks the first resolved IP, but fsock_get_contents() opens the actual connection using the original hostname. This causes a second DNS resolution and allows the connection to reach an address different from the one that was validated.
The response from the internal service is saved as an uploaded file and can be read through elFinder. This makes the demonstrated issue a non-blind SSRF.
The most accurate classification is:
- Vulnerability: SSRF protection bypass
- Exploitation technique: DNS rebinding
- Root cause: TOCTOU/double DNS resolution without IP pinning
Affected versions
Confirmed affected:
- elFinder 2.1.69, commit
8f2c3ffafcdd52cf4515f1eec172f4eee44552ad - the current
masterbranch inspected on 24 July 2026
The earliest affected version has not been determined.
Preconditions
The demonstrated readback path requires:
- PHP without the cURL extension (
curl_exec()unavailable); - access to the URL upload functionality;
- permission to upload a MIME type returned by the target service;
- network connectivity from the PHP process to the internal target;
- DNS resolution behavior compatible with low or zero TTL responses.
Authentication requirements depend on the application integrating the elFinder connector. The attached reproduction connector is intentionally minimal and unauthenticated.
Technical details
validate_address() resolves the supplied hostname with gethostbyname(), rejects loopback and private ranges, and stores the accepted address in $info['ip']:
https://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2587-L2655
get_remote_contents() then selects cURL when available, otherwise fsock_get_contents():
https://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2673-L2689
The cURL implementation correctly pins the validated IP using CURLOPT_RESOLVE:
https://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2709-L2736
However, the socket fallback ignores $info['ip'] and connects using $arr['host']:
$fp = fsockopen(
$ssl . $arr['host'],
$arr['port'],
$errno,
$errstr,
$connect_timeout
);
Relevant code:
https://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2762-L2866
Consequently, the address checked by validate_address() is not necessarily the address used by the socket connection.
The same problem is present after redirects: the redirect URL is validated, but the recursive socket request again connects using the hostname rather than the validated IP.
Proof of concept
A Docker-based reproduction is attached as poc.zip.
The laboratory:
- downloads the official elFinder 2.1.69 release during the build;
- verifies that the package version is 2.1.69;
- verifies that
curl_exec()is unavailable; - does not modify the elFinder core;
- runs a deterministic DNS server and an HTTP service bound only to loopback.
Build and run:
docker build -t elfinder-ssrf-repro poc
docker run --rm -p 127.0.0.1:8081:8081 elfinder-ssrf-repro
Open:
http://127.0.0.1:8081/poc/index.html
In the elFinder interface:
- Select "Upload".
- Paste the following URL:
http://rebind.test:9001/secret.txt
- Wait for the upload to complete.
- Open the resulting
secret.txtfile.
Observed content:
INTERNAL_SECRET_MANUAL
The PoC DNS server responds as follows:
first A resolution -> 203.0.113.10
subsequent A resolutions -> 127.0.0.1
Expected logs include:
DNS rebind.test type=1 answer=203.0.113.10 count=1
DNS rebind.test type=1 answer=127.0.0.1 count=2
GET /secret.txt
203.0.113.10 is used only as a deterministic laboratory address that passes the current private-address checks. The actual connection reaches 127.0.0.1:9001.
A redirect variant is also included:
http://rebind.test:9001/redirect-secret
It produces the same readable internal content.
Security impact
An attacker may cause the elFinder server to issue HTTP GET requests to services reachable from the PHP process, including:
- loopback services;
- private network services;
- link-local services where reachable;
- internal administrative or application endpoints.
Because the response body is saved and exposed through elFinder, sensitive internal information may be disclosed.
The demonstrated primitive is limited to HTTP GET requests and does not provide arbitrary request methods, bodies, or headers. Integrity and availability impact would depend on the behavior of reachable internal endpoints and are not claimed by this PoC.
Suggested CVSS
For an exposed connector matching the attached PoC:
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N
Base score: 8.6 High.
This assumes:
- no authentication is required by the connector;
- the internal service represents a separate security scope;
- disclosure may include highly sensitive internal data.
If authenticated user access is considered an inherent prerequisite, PR:L would result in a score of 7.7 High.
Suggested remediation
The socket connection should use the IP address returned by validate_address() rather than resolving the hostname again.
For HTTPS, the original hostname should still be used for:
- the HTTP
Hostheader; - TLS SNI;
- certificate hostname validation.
This may require replacing fsockopen() with a connection mechanism that supports connecting to the validated IP while explicitly configuring the TLS peer name.
Every redirect destination should be independently validated and its validated IP pinned to that specific connection.
Revalidating the hostname immediately before connecting is not sufficient because it would still leave a validation-to-connection DNS resolution gap.
As a temporary mitigation, URL uploads could be rejected when cURL is unavailable, or deployments could configure urlUploadFilter to reject URL uploads.
Additional observation
After a successful URL fetch, the upload path calls:
get_headers($url, true)
Relevant code:
https://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L3356-L3374
This performs another request using the original hostname without reusing the validated and pinned connection. It may therefore introduce an additional blind SSRF request, potentially even when the cURL download path is selected.
This additional request is not required for the attached non-blind fsock_get_contents() PoC. It should nevertheless be reviewed during remediation. Ideally, filename and response-header information should be collected from the already validated HTTP response instead of issuing a separate request.
Disclosure
Initial private report date: 24/07/2026
Reporter contact: Marco Lunardi - marcolunardi90@gmail.com
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 2.1.69"
},
"package": {
"ecosystem": "Packagist",
"name": "studio-42/elfinder"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2.1.70"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-81889"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2026-08-31T20:28:06Z",
"nvd_published_at": null,
"severity": "HIGH"
},
"details": "[poc.zip](https://github.com/user-attachments/files/30352020/poc.zip)\n## Summary\n\nelFinder 2.1.69 is vulnerable to a Server-Side Request Forgery (SSRF) protection bypass when PHP cURL is unavailable and URL uploads use the `fsock_get_contents()` fallback.\n\nAn attacker who can submit a URL for server-side upload can use an attacker-controlled DNS hostname that initially resolves to an allowed public IP address and subsequently resolves to a loopback or private IP address.\n\nThe URL validation checks the first resolved IP, but `fsock_get_contents()` opens the actual connection using the original hostname. This causes a second DNS resolution and allows the connection to reach an address different from the one that was validated.\n\nThe response from the internal service is saved as an uploaded file and can be read through elFinder. This makes the demonstrated issue a non-blind SSRF.\n\nThe most accurate classification is:\n\n* Vulnerability: SSRF protection bypass \n* Exploitation technique: DNS rebinding\n* Root cause: TOCTOU/double DNS resolution without IP pinning\n\n## Affected versions\n\nConfirmed affected:\n\n* elFinder 2.1.69, commit `8f2c3ffafcdd52cf4515f1eec172f4eee44552ad`\n* the current `master` branch inspected on 24 July 2026\n\nThe earliest affected version has not been determined.\n\n## Preconditions\n\nThe demonstrated readback path requires:\n\n* PHP without the cURL extension (`curl_exec()` unavailable);\n* access to the URL upload functionality;\n* permission to upload a MIME type returned by the target service;\n* network connectivity from the PHP process to the internal target;\n* DNS resolution behavior compatible with low or zero TTL responses.\n\nAuthentication requirements depend on the application integrating the elFinder connector. The attached reproduction connector is intentionally minimal and unauthenticated.\n\n## Technical details\n\n`validate_address()` resolves the supplied hostname with `gethostbyname()`, rejects loopback and private ranges, and stores the accepted address in `$info[\u0027ip\u0027]`:\n\nhttps://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2587-L2655\n\n`get_remote_contents()` then selects cURL when available, otherwise `fsock_get_contents()`:\n\nhttps://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2673-L2689\n\nThe cURL implementation correctly pins the validated IP using `CURLOPT_RESOLVE`:\n\nhttps://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2709-L2736\n\nHowever, the socket fallback ignores `$info[\u0027ip\u0027]` and connects using `$arr[\u0027host\u0027]`:\n\n```php\n$fp = fsockopen(\n $ssl . $arr[\u0027host\u0027],\n $arr[\u0027port\u0027],\n $errno,\n $errstr,\n $connect_timeout\n);\n```\n\nRelevant code:\n\nhttps://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L2762-L2866\n\nConsequently, the address checked by `validate_address()` is not necessarily the address used by the socket connection.\n\nThe same problem is present after redirects: the redirect URL is validated, but the recursive socket request again connects using the hostname rather than the validated IP.\n\n## Proof of concept\n\nA Docker-based reproduction is attached as `poc.zip`.\n\nThe laboratory:\n\n* downloads the official elFinder 2.1.69 release during the build;\n* verifies that the package version is 2.1.69;\n* verifies that `curl_exec()` is unavailable;\n* does not modify the elFinder core;\n* runs a deterministic DNS server and an HTTP service bound only to loopback.\n\nBuild and run:\n\n```sh\ndocker build -t elfinder-ssrf-repro poc\ndocker run --rm -p 127.0.0.1:8081:8081 elfinder-ssrf-repro\n```\n\nOpen:\n\n```text\nhttp://127.0.0.1:8081/poc/index.html\n```\n\nIn the elFinder interface:\n\n1. Select \"Upload\".\n2. Paste the following URL:\n\n```text\nhttp://rebind.test:9001/secret.txt\n```\n\n3. Wait for the upload to complete.\n4. Open the resulting `secret.txt` file.\n\nObserved content:\n\n```text\nINTERNAL_SECRET_MANUAL\n```\n\nThe PoC DNS server responds as follows:\n\n```text\nfirst A resolution -\u003e 203.0.113.10\nsubsequent A resolutions -\u003e 127.0.0.1\n```\n\nExpected logs include:\n\n```text\nDNS rebind.test type=1 answer=203.0.113.10 count=1\nDNS rebind.test type=1 answer=127.0.0.1 count=2\nGET /secret.txt\n```\n\n`203.0.113.10` is used only as a deterministic laboratory address that passes the current private-address checks. The actual connection reaches `127.0.0.1:9001`.\n\nA redirect variant is also included:\n\n```text\nhttp://rebind.test:9001/redirect-secret\n```\n\nIt produces the same readable internal content.\n\n## Security impact\n\nAn attacker may cause the elFinder server to issue HTTP GET requests to services reachable from the PHP process, including:\n\n* loopback services;\n* private network services;\n* link-local services where reachable;\n* internal administrative or application endpoints.\n\nBecause the response body is saved and exposed through elFinder, sensitive internal information may be disclosed.\n\nThe demonstrated primitive is limited to HTTP GET requests and does not provide arbitrary request methods, bodies, or headers. Integrity and availability impact would depend on the behavior of reachable internal endpoints and are not claimed by this PoC.\n\n## Suggested CVSS\n\nFor an exposed connector matching the attached PoC:\n\n```text\nCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N\n```\n\nBase score: 8.6 High.\n\nThis assumes:\n\n* no authentication is required by the connector;\n* the internal service represents a separate security scope;\n* disclosure may include highly sensitive internal data.\n\nIf authenticated user access is considered an inherent prerequisite, `PR:L` would result in a score of 7.7 High.\n\n## Suggested remediation\n\nThe socket connection should use the IP address returned by `validate_address()` rather than resolving the hostname again.\n\nFor HTTPS, the original hostname should still be used for:\n\n* the HTTP `Host` header;\n* TLS SNI;\n* certificate hostname validation.\n\nThis may require replacing `fsockopen()` with a connection mechanism that supports connecting to the validated IP while explicitly configuring the TLS peer name.\n\nEvery redirect destination should be independently validated and its validated IP pinned to that specific connection.\n\nRevalidating the hostname immediately before connecting is not sufficient because it would still leave a validation-to-connection DNS resolution gap.\n\nAs a temporary mitigation, URL uploads could be rejected when cURL is unavailable, or deployments could configure `urlUploadFilter` to reject URL uploads.\n\n## Additional observation\n\nAfter a successful URL fetch, the upload path calls:\n\n```php\nget_headers($url, true)\n```\n\nRelevant code:\n\nhttps://github.com/Studio-42/elFinder/blob/2.1.69/php/elFinder.class.php#L3356-L3374\n\nThis performs another request using the original hostname without reusing the validated and pinned connection. It may therefore introduce an additional blind SSRF request, potentially even when the cURL download path is selected.\n\nThis additional request is not required for the attached non-blind `fsock_get_contents()` PoC. It should nevertheless be reviewed during remediation. Ideally, filename and response-header information should be collected from the already validated HTTP response instead of issuing a separate request.\n\n## Disclosure\n\nInitial private report date: `24/07/2026`\n\nReporter contact: `Marco Lunardi - marcolunardi90@gmail.com`",
"id": "GHSA-8x3q-jpjh-qh5c",
"modified": "2026-08-31T20:28:06Z",
"published": "2026-08-31T20:28:06Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/Studio-42/elFinder/security/advisories/GHSA-8x3q-jpjh-qh5c"
},
{
"type": "WEB",
"url": "https://github.com/Studio-42/elFinder/commit/191372c1bbebbd36fb55af79a84b9984861390ff"
},
{
"type": "WEB",
"url": "https://github.com/Studio-42/elFinder/commit/6d997386cd0f1abab4706c220b46b0aea0ecff51"
},
{
"type": "PACKAGE",
"url": "https://github.com/Studio-42/elFinder"
},
{
"type": "WEB",
"url": "https://github.com/Studio-42/elFinder/releases/tag/2.1.70"
}
],
"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": "elFinder: SSRF protection bypass via DNS rebinding in the `fsock_get_contents()` fallback"
}
GHSA-8X5V-CPV7-8JJP
Vulnerability from github – Published: 2026-08-04 19:59 – Updated: 2026-08-04 19:59Summary
Open WebUI fetches user-supplied URLs on the server for RAG URL ingestion, URL-to-markdown conversion and web-search content retrieval, and decides whether a destination is allowed by asking whether its IP address is globally routable. That test operates on the literal IPv6 address and does not look at the IPv4 address embedded inside it. On a deployment whose network has a NAT64 gateway, any verified user can wrap an internal or cloud-metadata IPv4 address in the NAT64 well-known prefix, pass the filter, and receive the internal response body back through the API.
Preconditions
- Any verified (authenticated) user account. No admin role, no elevated permission.
- Default configuration:
ENABLE_LOCAL_WEB_FETCHoff, the defaultWEB_FETCH_FILTER_LISTmetadata blocklist in place. Neither prevents this, because the blocklist matches hostname strings and the NAT64 literal is not one of them. - The deployment's network must provide NAT64 translation for the well-known
64:ff9b::/96prefix, which is the common default on IPv6-only and dual-stack cloud and Kubernetes networks. - Deployments on IPv4-only networks, or on any network without a NAT64 gateway, are not affected: the address has nowhere to route.
Impact
On an affected network a low-privilege user can read GET responses from services the server can reach but the internet cannot: cloud instance metadata including IAM role credentials, loopback-bound admin surfaces, and internal APIs in the same VPC or cluster. The response body is returned to the caller, so this is full-read, not blind. Exploitation is not universal, it depends entirely on the deployment's network providing NAT64 translation, which is why the score carries high attack complexity. Deployments without NAT64 lose nothing here.
Fix
Fixed in v0.11.0 by commit 1717b493d. Address classification now unwraps the IPv4 embedded in IPv6 transition encodings before deciding whether a destination is global, and applies that at all three checkpoints. NAT64-wrapped public destinations continue to work. Upgrading to v0.11.0 fully resolves the issue with no configuration change.
Root cause
backend/open_webui/retrieval/web/utils.py—validate_url(), the pre-fetch check on the submitted URL.backend/open_webui/retrieval/web/utils.py—_ssrf_safe_new_conn()and_SSRFSafeResolver, the connect-time re-checks that defeat DNS rebinding.
All three decided reachability from ipaddress.ip_address(ip).is_global applied to the literal address. That predicate answers whether an IPv6 address sits in globally-routable space, which is a different question from where the packet actually ends up once a transition gateway translates it. The NAT64 well-known prefix is by design a global prefix carrying an arbitrary IPv4 destination, so an internal target wrapped in it satisfies the check while reaching exactly what the check exists to prevent. Because the same predicate backed the connect-time re-checks, no later layer caught it either. The fix inspects every standardized transition encoding rather than only the NAT64 prefix, since the same reasoning error applies to each of them.
Proof of concept
Against the real POST /api/v1/retrieval/process/web flow on v0.10.2 as an authenticated user, with internal HTTP services returning a marker string. The plain forms are rejected with HTTP 400:
http://169.254.169.254/latest/meta-data/ -> 400
http://127.0.0.1/ -> 400
http://[::ffff:169.254.169.254]/ -> 400
http://metadata.google.internal/ -> 400
The NAT64 encodings of the same targets are accepted, and the response body is returned in the content field:
http://[64:ff9b::a9fe:a9fe]/latest/meta-data/iam/security-credentials/ -> 200, marker returned
http://[64:ff9b::7f00:1]/admin/internal-status -> 200, marker returned
NAT64 translation was modelled by binding the translated addresses locally rather than by routing through a real NAT64 gateway; everything else, including the request flow and the validation code, is the unmodified v0.10.2 path. After the fix both URLs return 400 while http://[64:ff9b::808:808]/ (8.8.8.8, public) still returns 200, confirming no over-blocking.
Credits
- tonghuaroot — reported the transition-form gap in the address classification and supplied the fix approach.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "open-webui"
},
"ranges": [
{
"events": [
{
"introduced": "0.9.0"
},
{
"fixed": "0.11.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-70485"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2026-08-04T19:59:36Z",
"nvd_published_at": null,
"severity": "HIGH"
},
"details": "## Summary\n\nOpen WebUI fetches user-supplied URLs on the server for RAG URL ingestion, URL-to-markdown conversion and web-search content retrieval, and decides whether a destination is allowed by asking whether its IP address is globally routable. That test operates on the literal IPv6 address and does not look at the IPv4 address embedded inside it. On a deployment whose network has a NAT64 gateway, any verified user can wrap an internal or cloud-metadata IPv4 address in the NAT64 well-known prefix, pass the filter, and receive the internal response body back through the API.\n\n## Preconditions\n\n- Any verified (authenticated) user account. No admin role, no elevated permission.\n- Default configuration: `ENABLE_LOCAL_WEB_FETCH` off, the default `WEB_FETCH_FILTER_LIST` metadata blocklist in place. Neither prevents this, because the blocklist matches hostname strings and the NAT64 literal is not one of them.\n- The deployment\u0027s network must provide NAT64 translation for the well-known `64:ff9b::/96` prefix, which is the common default on IPv6-only and dual-stack cloud and Kubernetes networks.\n- Deployments on IPv4-only networks, or on any network without a NAT64 gateway, are not affected: the address has nowhere to route.\n\n## Impact\n\nOn an affected network a low-privilege user can read GET responses from services the server can reach but the internet cannot: cloud instance metadata including IAM role credentials, loopback-bound admin surfaces, and internal APIs in the same VPC or cluster. The response body is returned to the caller, so this is full-read, not blind. Exploitation is not universal, it depends entirely on the deployment\u0027s network providing NAT64 translation, which is why the score carries high attack complexity. Deployments without NAT64 lose nothing here.\n\n## Fix\n\nFixed in v0.11.0 by commit `1717b493d`. Address classification now unwraps the IPv4 embedded in IPv6 transition encodings before deciding whether a destination is global, and applies that at all three checkpoints. NAT64-wrapped public destinations continue to work. Upgrading to v0.11.0 fully resolves the issue with no configuration change.\n\n## Root cause\n\n- `backend/open_webui/retrieval/web/utils.py` \u2014 `validate_url()`, the pre-fetch check on the submitted URL.\n- `backend/open_webui/retrieval/web/utils.py` \u2014 `_ssrf_safe_new_conn()` and `_SSRFSafeResolver`, the connect-time re-checks that defeat DNS rebinding.\n\nAll three decided reachability from `ipaddress.ip_address(ip).is_global` applied to the literal address. That predicate answers whether an IPv6 address sits in globally-routable space, which is a different question from where the packet actually ends up once a transition gateway translates it. The NAT64 well-known prefix is by design a global prefix carrying an arbitrary IPv4 destination, so an internal target wrapped in it satisfies the check while reaching exactly what the check exists to prevent. Because the same predicate backed the connect-time re-checks, no later layer caught it either. The fix inspects every standardized transition encoding rather than only the NAT64 prefix, since the same reasoning error applies to each of them.\n\n## Proof of concept\n\nAgainst the real `POST /api/v1/retrieval/process/web` flow on v0.10.2 as an authenticated user, with internal HTTP services returning a marker string. The plain forms are rejected with HTTP 400:\n\n```\nhttp://169.254.169.254/latest/meta-data/ -\u003e 400\nhttp://127.0.0.1/ -\u003e 400\nhttp://[::ffff:169.254.169.254]/ -\u003e 400\nhttp://metadata.google.internal/ -\u003e 400\n```\n\nThe NAT64 encodings of the same targets are accepted, and the response body is returned in the `content` field:\n\n```\nhttp://[64:ff9b::a9fe:a9fe]/latest/meta-data/iam/security-credentials/ -\u003e 200, marker returned\nhttp://[64:ff9b::7f00:1]/admin/internal-status -\u003e 200, marker returned\n```\n\nNAT64 translation was modelled by binding the translated addresses locally rather than by routing through a real NAT64 gateway; everything else, including the request flow and the validation code, is the unmodified v0.10.2 path. After the fix both URLs return 400 while `http://[64:ff9b::808:808]/` (8.8.8.8, public) still returns 200, confirming no over-blocking.\n\n## Credits\n\n- tonghuaroot \u2014 reported the transition-form gap in the address classification and supplied the fix approach.",
"id": "GHSA-8x5v-cpv7-8jjp",
"modified": "2026-08-04T19:59:36Z",
"published": "2026-08-04T19:59:36Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/open-webui/open-webui/security/advisories/GHSA-8x5v-cpv7-8jjp"
},
{
"type": "WEB",
"url": "https://github.com/open-webui/open-webui/commit/1717b493d83c86afa82aa8bc50139250852dd2f3"
},
{
"type": "PACKAGE",
"url": "https://github.com/open-webui/open-webui"
},
{
"type": "WEB",
"url": "https://github.com/open-webui/open-webui/releases/tag/v0.11.0"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:L/A:N",
"type": "CVSS_V3"
}
],
"summary": "Open WebUI: Any authenticated user can reach internal services and cloud metadata via NAT64-encoded URLs"
}
GHSA-8X9J-8H3W-5CXQ
Vulnerability from github – Published: 2026-09-26 15:31 – Updated: 2026-09-26 15:31Kyverno before 1.19.1 is vulnerable to server-side request forgery. The default egress blocklist (169.254.169.254, 169.254.169.253, metadata.google.internal, 127.0.0.0/8, ::1/128) and the scoped-token control were wired only into the new CEL http.Get/Post library and were never applied to the legacy apiCall service executor (pkg/engine/apicall/executor.go) or to the GlobalContextEntry external-API path, which handle every non-CEL context[].apiCall.service call. Because these paths use a plain net/http client with no egress filtering and no validation of the configured service URL, a ClusterPolicy or GlobalContextEntry author — or, where a deployed policy templates the service URL from the admission resource, a lower-privileged resource submitter — can cause Kyverno to issue GET/POST requests to an arbitrary host, including the cloud metadata endpoint, loopback, and any in-cluster service, reading cloud instance credentials and reaching internal endpoints with Kyverno's network position. The executor also unconditionally attaches Kyverno's projected ServiceAccount token to the attacker-chosen destination; the token is audience-scoped, limiting its replay value. Fixed in 1.19.1.
{
"affected": [],
"aliases": [
"CVE-2026-100705"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-26T14:16:55Z",
"severity": "HIGH"
},
"details": "Kyverno before 1.19.1 is vulnerable to server-side request forgery. The default egress blocklist (169.254.169.254, 169.254.169.253, metadata.google.internal, 127.0.0.0/8, ::1/128) and the scoped-token control were wired only into the new CEL http.Get/Post library and were never applied to the legacy apiCall service executor (pkg/engine/apicall/executor.go) or to the GlobalContextEntry external-API path, which handle every non-CEL context[].apiCall.service call. Because these paths use a plain net/http client with no egress filtering and no validation of the configured service URL, a ClusterPolicy or GlobalContextEntry author \u2014 or, where a deployed policy templates the service URL from the admission resource, a lower-privileged resource submitter \u2014 can cause Kyverno to issue GET/POST requests to an arbitrary host, including the cloud metadata endpoint, loopback, and any in-cluster service, reading cloud instance credentials and reaching internal endpoints with Kyverno\u0027s network position. The executor also unconditionally attaches Kyverno\u0027s projected ServiceAccount token to the attacker-chosen destination; the token is audience-scoped, limiting its replay value. Fixed in 1.19.1.",
"id": "GHSA-8x9j-8h3w-5cxq",
"modified": "2026-09-26T15:31:23Z",
"published": "2026-09-26T15:31:22Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/kyverno/kyverno/security/advisories/GHSA-q825-p383-r9v5"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-100705"
},
{
"type": "WEB",
"url": "https://www.vulncheck.com/advisories/kyverno-before-1.19.1-ssrf-via-legacy-apicall-service-executor"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:C/C:H/I:L/A:N",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:H/UI:N/VC:H/VI:L/VA:N/SC:H/SI:L/SA:N/E:X/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-8XCX-2PHJ-G7JH
Vulnerability from github – Published: 2026-09-01 21:31 – Updated: 2026-09-01 21:31A vulnerability in the web-based management interface of AOS-CX could allow an authenticated remote attacker to conduct a server-side request forgery (SSRF) attack. A successful exploit allows an attacker to enumerate information about the internal structure of the AOS-CX host, leading to potential disclosure and limited modification of sensitive information.
{
"affected": [],
"aliases": [
"CVE-2026-73757"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-01T21:18:42Z",
"severity": "MODERATE"
},
"details": "A vulnerability in the web-based management interface of AOS-CX could allow an authenticated remote attacker to conduct a server-side request forgery (SSRF) attack. A successful exploit allows an attacker to enumerate information about the internal structure of the AOS-CX host, leading to potential disclosure and limited modification of sensitive information.",
"id": "GHSA-8xcx-2phj-g7jh",
"modified": "2026-09-01T21:31:51Z",
"published": "2026-09-01T21:31:51Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-73757"
},
{
"type": "WEB",
"url": "https://support.hpe.com/hpesc/public/docDisplay?docId=hpesbnw05134en_us\u0026docLocale=en_US"
}
],
"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"
}
]
}
GHSA-8XMV-482H-9XQG
Vulnerability from github – Published: 2022-05-24 19:09 – Updated: 2022-05-24 19:09IBM Jazz Foundation products are 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: 192434.
{
"affected": [],
"aliases": [
"CVE-2020-4974"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2021-07-28T13:15:00Z",
"severity": "MODERATE"
},
"details": "IBM Jazz Foundation products are 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: 192434.",
"id": "GHSA-8xmv-482h-9xqg",
"modified": "2022-05-24T19:09:17Z",
"published": "2022-05-24T19:09:17Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2020-4974"
},
{
"type": "WEB",
"url": "https://exchange.xforce.ibmcloud.com/vulnerabilities/192434"
},
{
"type": "WEB",
"url": "https://www.ibm.com/support/pages/node/6475919"
}
],
"schema_version": "1.4.0",
"severity": []
}
GHSA-8XPW-Q9JP-X2C7
Vulnerability from github – Published: 2022-07-15 00:00 – Updated: 2022-07-22 00:00Best Practical RT for Incident Response (RTIR) before 4.0.3 and 5.x before 5.0.3 allows SSRF via the whois lookup tool.
{
"affected": [],
"aliases": [
"CVE-2022-25800"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2022-07-14T12:15:00Z",
"severity": "CRITICAL"
},
"details": "Best Practical RT for Incident Response (RTIR) before 4.0.3 and 5.x before 5.0.3 allows SSRF via the whois lookup tool.",
"id": "GHSA-8xpw-q9jp-x2c7",
"modified": "2022-07-22T00:00:33Z",
"published": "2022-07-15T00:00:21Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2022-25800"
},
{
"type": "WEB",
"url": "https://docs.bestpractical.com/release-notes/rtir/4.0.3"
},
{
"type": "WEB",
"url": "https://docs.bestpractical.com/release-notes/rtir/5.0.3"
},
{
"type": "WEB",
"url": "https://docs.bestpractical.com/release-notes/rtir/index.html"
}
],
"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"
}
]
}
GHSA-8XQ3-W9FX-74RV
Vulnerability from github – Published: 2025-07-28 16:41 – Updated: 2025-08-01 18:36Description
The lookup function takes a user address for checking accounts as a feature, however, as per the ActivityPub spec (https://www.w3.org/TR/activitypub/#security-considerations), on the security considerations section at B.3, access to Localhost services should be prevented while running in production. The library does not prevent Localhost access (neither does it prevent LAN addresses such as 192.168.x.x) , thus is not safe for use in production by ActivityPub applications. The only check for localhost is done for selecting between HTTP and HTTPS protocols, and it is done by testing for a host that starts with the string “localhost” and ends with a port. Anything else (such as “127.0.0.1” or “localhost:1234/abc”) would not be considered localhost for this test.
In addition, the way that the function determines the host, makes it possible to access any path in the host, not only “/.well-known/...” paths:
if (address.indexOf('://') > -1) {
// other uri format
host = address.replace(/ /g,'').split('/')[2];
} else {
// useraddress
host = address.replace(/ /g,'').split('@')[1];
}
var uri_index = 0; // track which URIS we've tried already
var protocol = 'https'; // we use https by default
if (self.__isLocalhost(host)) {
protocol = 'http';
}
function __buildURL() {
var uri = '';
if (! address.split('://')[1]) {
// the URI has not been defined, default to acct
uri = 'acct:';
}
return protocol + '://' + host + '/.well-known/' +URIS[uri_index] + '?resource=' + uri + address;
}
If the address is in the format of a user address (user@host.com), the host will be anything after the first found @ symbol. Since no other test is done, an adversary may pass a specially crafted address such as user@localhost:7000/admin/restricted_page? and reach pages that would normally be out of reach. In this example, the code would treat localhost:7000/admin/restricted_page? as the host, and the created URL would be https://localhost:7000/admin/restricted_page?/.well-known/webfinger?resource=acct:use r@localhost:7000/admin/restricted_page?. A server listening on localhost:7000 will then parse the request as a GET request for the page /admin/restricted_page with the query string /.well-known/webfinger?resource=acct:user@localhost:7000/admin/restricted_page?.
PoC and Steps to reproduce
This PoC assumes that there is a server on the machine listening on port 3000, which receives requests for WebFinger lookups on the address /api/v1/search_user, and then calls the lookup function in webfinger.js with the user passed as an argument. For the sake of the example we assume that the server configured webfinger.js with tls_only=false.
- Activate a local HTTP server listening to port 1234 with a “secret.txt” file:
python3 -m http.server 1234
- Run the following command:
curl
"http://localhost:3000/api/v1/search_user?search=user@localhost:1234/secret.txt
?"
- View the console of the Python’s HTTP server and see that a request for a “secret.txt?/.well-known/webfinger?resource=acct:user@localhost:1234/secret.txt ?” file was performed. This proves that we can redirect the URL to any domain and path we choose, including localhost and the internal LAN.
Impact
Due to this issue, any user can cause a server using the library to send GET requests with controlled host, path and port in an attempt to query services running on the instance’s host or local network, and attempt to execute a Blind-SSRF gadget in hope of targeting a known vulnerable local service running on the victim’s machine.
References
The vulnerability was discovered by Ori Hollander of the JFrog Vulnerability Research team.
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 2.8.0"
},
"package": {
"ecosystem": "npm",
"name": "webfinger.js"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2.8.1"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2025-54590"
],
"database_specific": {
"cwe_ids": [
"CWE-918"
],
"github_reviewed": true,
"github_reviewed_at": "2025-07-28T16:41:06Z",
"nvd_published_at": "2025-08-01T18:15:55Z",
"severity": "MODERATE"
},
"details": "### Description\nThe lookup function takes a user address for checking accounts as a feature, however, as per\nthe ActivityPub spec (https://www.w3.org/TR/activitypub/#security-considerations), on the\nsecurity considerations section at B.3, access to Localhost services should be prevented while\nrunning in production. The library does not prevent Localhost access (neither does it prevent\nLAN addresses such as 192.168.x.x) , thus is not safe for use in production by ActivityPub\napplications. The only check for localhost is done for selecting between HTTP and HTTPS\nprotocols, and it is done by testing for a host that starts with the string \u201clocalhost\u201d and ends with\na port. Anything else (such as \u201c127.0.0.1\u201d or \u201clocalhost:1234/abc\u201d) would not be considered\nlocalhost for this test.\n\nIn addition, the way that the function determines the host, makes it possible to access any path\nin the host, not only \u201c/.well-known/...\u201d paths:\n\n```javascript\nif (address.indexOf(\u0027://\u0027) \u003e -1) {\n // other uri format\n host = address.replace(/ /g,\u0027\u0027).split(\u0027/\u0027)[2];\n} else {\n // useraddress\n host = address.replace(/ /g,\u0027\u0027).split(\u0027@\u0027)[1];\n}\n\nvar uri_index = 0; // track which URIS we\u0027ve tried already\nvar protocol = \u0027https\u0027; // we use https by default\n\nif (self.__isLocalhost(host)) {\n protocol = \u0027http\u0027;\n}\n\nfunction __buildURL() {\n var uri = \u0027\u0027;\n if (! address.split(\u0027://\u0027)[1]) {\n // the URI has not been defined, default to acct\n uri = \u0027acct:\u0027;\n }\n return protocol + \u0027://\u0027 + host + \u0027/.well-known/\u0027 +URIS[uri_index] + \u0027?resource=\u0027 + uri + address;\n}\n```\n\nIf the address is in the format of a user address (user@host.com), the host will be anything\nafter the first found @ symbol. Since no other test is done, an adversary may pass a specially\ncrafted address such as user@localhost:7000/admin/restricted_page? and reach pages that\nwould normally be out of reach. In this example, the code would treat\nlocalhost:7000/admin/restricted_page? as the host, and the created URL would be\nhttps://localhost:7000/admin/restricted_page?/.well-known/webfinger?resource=acct:use\nr@localhost:7000/admin/restricted_page?. A server listening on localhost:7000 will then\nparse the request as a GET request for the page /admin/restricted_page with the query string\n/.well-known/webfinger?resource=acct:user@localhost:7000/admin/restricted_page?.\n\n### PoC and Steps to reproduce\nThis PoC assumes that there is a server on the machine listening on port 3000, which receives\nrequests for WebFinger lookups on the address /api/v1/search_user, and then calls the lookup\nfunction in webfinger.js with the user passed as an argument. For the sake of the example we\nassume that the server configured webfinger.js with tls_only=false.\n\n\n1. Activate a local HTTP server listening to port 1234 with a \u201csecret.txt\u201d file:\n\n```\npython3 -m http.server 1234\n```\n\n2. Run the following command:\n\n```\ncurl\n\"http://localhost:3000/api/v1/search_user?search=user@localhost:1234/secret.txt\n?\"\n```\n\n3. View the console of the Python\u2019s HTTP server and see that a request for a\n\u201csecret.txt?/.well-known/webfinger?resource=acct:user@localhost:1234/secret.txt\n?\u201d file was performed.\nThis proves that we can redirect the URL to any domain and path we choose, including\nlocalhost and the internal LAN.\n\n\n### Impact\nDue to this issue, any user can cause a server using the library to send GET requests with\ncontrolled host, path and port in an attempt to query services running on the instance\u2019s host or\nlocal network, and attempt to execute a Blind-SSRF gadget in hope of targeting a known\nvulnerable local service running on the victim\u2019s machine.\n\n\n### References\nThe vulnerability was discovered by Ori Hollander of the JFrog Vulnerability Research team.",
"id": "GHSA-8xq3-w9fx-74rv",
"modified": "2025-08-01T18:36:22Z",
"published": "2025-07-28T16:41:06Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/silverbucket/webfinger.js/security/advisories/GHSA-8xq3-w9fx-74rv"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-54590"
},
{
"type": "WEB",
"url": "https://github.com/silverbucket/webfinger.js/commit/b5f2f2c957297d25f4d76072963fccaee2e3095a"
},
{
"type": "PACKAGE",
"url": "https://github.com/silverbucket/webfinger.js"
},
{
"type": "WEB",
"url": "https://github.com/silverbucket/webfinger.js/releases/tag/v2.8.1"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N",
"type": "CVSS_V4"
}
],
"summary": "webfinger.js Blind SSRF Vulnerability"
}
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.