BentoML is Vulnerable to an SSRF Attack Through File Upload Processing

BentoML is a Python library for building online serving systems optimized for AI apps and model inference. In versions 1.4.0 until 1.4.19, the file upload processing system contains an SSRF vulnerability that allows unauthenticated remote attackers to force the server to make arbitrary HTTP requests. The vulnerability stems from the multipart form data and JSON request handlers, which automatically download files from user-provided URLs without validating whether those URLs point to internal network addresses, cloud metadata endpoints, or other restricted resources. The documentation explicitly promotes this URL-based file upload feature, making it an intended design that exposes all deployed services to SSRF attacks by default. Version 1.4.19 contains a patch for the issue.

GLPI has overly permissive URL verification

GLPI is a Free Asset and IT Management Software package, Data center management, ITIL Service Desk, licenses tracking and software auditing. In versions 0.84 through 10.0.18, usage of RSS feeds or external calendars when planning is subject to SSRF exploit. The previous security patches provided since GLPI 10.0.4 were not robust enough for certain specific cases. This is fixed in version 10.0.19.

2FAuth vulnerable to Server Side Request Forgery + URI validation bypass in 2fauth /api/v1/twofaccounts/preview

2FAuth is a web app to manage Two-Factor Authentication (2FA) accounts and generate their security codes. Two interconnected vulnerabilities exist in version 5.4.1 a SSRF and URI validation bypass issue. The endpoint at POST /api/v1/twofaccounts/preview allows setting a remote URI to retrieve the image of a 2fa site. By abusing this functionality, it is possible to force the application to make a GET request to an arbitrary URL, whose content will be stored in an image file in the server if it looks like an image. Additionally, the library does some basic validation on the URI, attempting to filter our URIs which do not have an image extension. However, this can be easily bypassed by appending the string `#.svg` to the URI. The combination of these two issues allows an attacker to retrieve URIs accessible from the application, as long as their content type is text based. If not, the request is still sent, but the response is not reflected to the attacker. Version 5.4.1 fixes the issues.

SSRF in comfyanonymous/comfyui

comfyanonymous/comfyui version v0.2.4 suffers from a non-blind Server-Side Request Forgery (SSRF) vulnerability. This vulnerability can be exploited by combining the REST APIs `POST /internal/models/download` and `GET /view`, allowing attackers to abuse the victim server's credentials to access unauthorized web resources.

Applio allows SSRF and file write in model_download.py

Applio is a voice conversion tool. Versions 3.2.7 and prior are vulnerable to server-side request forgery (SSRF) and file write in `model_download.py` (line 156 in 3.2.7). The blind SSRF allows for sending requests on behalf of Applio server and can be leveraged to probe for other vulnerabilities on the server itself or on other back-end systems on the internal network, that the Applio server can reach. The blind SSRF can also be coupled with the an arbitrary file read (e.g., CVE-2025-27784) to read files from hosts on the internal network, that the Applio server can reach, which would make it a full SSRF. The file write allows for writing files on the server, which can be coupled with other vulnerabilities, for example an unsafe deserialization, to achieve remote code execution on the Applio server. As of time of publication, no known patches are available.

Applio allows SSRF and file write in model_download.py

Applio is a voice conversion tool. Versions 3.2.7 and prior are vulnerable to server-side request forgery (SSRF) and file write in `model_download.py` (line 143 in 3.2.7). The blind SSRF allows for sending requests on behalf of Applio server and can be leveraged to probe for other vulnerabilities on the server itself or on other back-end systems on the internal network, that the Applio server can reach. The file write allows for writing files on the server, which can be coupled with other vulnerabilities, for example an unsafe deserialization, to achieve remote code execution on the Applio server. As of time of publication, no known patches are available.

Applio allows SSRF and file write in model_download.py

Applio is a voice conversion tool. Versions 3.2.7 and prior are vulnerable to server-side request forgery (SSRF) and file write in `model_download.py` (line 240 in 3.2.7). The blind SSRF allows for sending requests on behalf of Applio server and can be leveraged to probe for other vulnerabilities on the server itself or on other back-end systems on the internal network, that the Applio server can reach. The blind SSRF can also be coupled with the arbitrary file read CVE-2025-27784 to read files from hosts on the internal network, that the Applio server can reach, which would make it a full SSRF. The file write allows for writing files on the server, which can be coupled with other vulnerabilities, for example an unsafe deserialization, to achieve remote code execution on the Applio server. As of time of publication, no known patches are available.

Applio allows SSRF and file write in model_download.py

Applio is a voice conversion tool. Versions 3.2.7 and prior are vulnerable to server-side request forgery (SSRF) in `model_download.py` (line 195 in 3.2.7). The blind SSRF allows for sending requests on behalf of Applio server and can be leveraged to probe for other vulnerabilities on the server itself or on other back-end systems on the internal network, that the Applio server can reach. The blind SSRF can also be coupled with a arbitrary file read (e.g., CVE-2025-27784) to read files from hosts on the internal network, that the Applio server can reach, which would make it a full SSRF. As of time of publication, no known patches are available.

Server-Side Request Forgery (SSRF) in binary-husky/gpt_academic

A Server-Side Request Forgery (SSRF) vulnerability exists in binary-husky/gpt_academic version git 310122f. The application has a functionality to download papers from arxiv.org, but the URL validation is incomplete. An attacker can exploit this vulnerability to make the application access any URL, including internal services, and read the response. This can be used to access data that are only accessible from the server, such as AWS metadata credentials, and can escalate local exploits to network-based attacks.

Server Side Request Forgery in lm-sys/fastchat

A Server-Side Request Forgery (SSRF) vulnerability was identified in the lm-sys/fastchat web server, specifically in the affected version git 2c68a13. This vulnerability allows an attacker to access internal server resources and data that are otherwise inaccessible, such as AWS metadata credentials.

SSRF in GravityZone Console via DNS Truncation (VA-12634)

A server-side request forgery (SSRF) vulnerability in Bitdefender GravityZone Console allows an attacker to bypass input validation logic using leading characters in DNS requests. Paired with other potential vulnerabilities, this bypass could be used for execution of third party code. This issue affects GravityZone Console: before 6.41.2.1.

SSRF in POST /worker_generate_stream API endpoint in lm-sys/fastchat

A Server-Side Request Forgery (SSRF) vulnerability exists in the POST /worker_generate_stream API endpoint of the Controller API Server in lm-sys/fastchat, as of commit e208d5677c6837d590b81cb03847c0b9de100765. This vulnerability allows attackers to exploit the victim controller API server's credentials to perform unauthorized web actions or access unauthorized web resources by combining it with the POST /register_worker endpoint.

SSRF Vulnerability in gradio-app/gradio

An SSRF vulnerability exists in the gradio-app/gradio due to insufficient validation of user-supplied URLs in the `/proxy` route. Attackers can exploit this vulnerability by manipulating the `self.replica_urls` set through the `X-Direct-Url` header in requests to the `/` and `/config` routes, allowing the addition of arbitrary URLs for proxying. This flaw enables unauthorized proxying of requests and potential access to internal endpoints within the Hugging Face space. The issue arises from the application's inadequate checking of safe URLs in the `build_proxy_request` function.

Server-Side Request Forgery in lm-sys/fastchat

A Server-Side Request Forgery (SSRF) vulnerability exists in lm-sys/fastchat version 0.2.36. The vulnerability is present in the `/queue/join?` endpoint, where insufficient validation of the path parameter allows an attacker to send crafted requests. This can lead to unauthorized access to internal networks or the AWS metadata endpoint, potentially exposing sensitive data and compromising internal servers.

Server-Side Request Forgery (SSRF)

Server-Side Request Forgery (SSRF) in Citrix SD-WAN Standard/Premium Editions on or after 11.4.0 and before 11.4.4.46 allows an attacker to disclose limited information from the appliance via Access to management IP.

Cisco Unified Intelligence Center Server-Side Request Forgery Vulnerability

A vulnerability in the web-based management interface of Cisco Unified Intelligence Center could allow an unauthenticated, remote attacker to conduct a server-side request forgery (SSRF) attack through an affected device. This vulnerability is due to improper input validation for specific HTTP requests. An attacker could exploit this vulnerability by sending a crafted HTTP request to an affected device. A successful exploit could allow the attacker to send arbitrary network requests that are sourced from the affected device.

In Eclipse GlassFish since version 6.2.5 it is possible to perform a Server Side Request Forgery attack in specific endpoints.

SSRF Vulnerability in RequestsToolkit in langchain-ai/langchain

A Server-Side Request Forgery (SSRF) vulnerability exists in the RequestsToolkit component of the langchain-community package (specifically, langchain_community.agent_toolkits.openapi.toolkit.RequestsToolkit) in langchain-ai/langchain version 0.0.27. This vulnerability occurs because the toolkit does not enforce restrictions on requests to remote internet addresses, allowing it to also access local addresses. As a result, an attacker could exploit this flaw to perform port scans, access local services, retrieve instance metadata from cloud environments (e.g., Azure, AWS), and interact with servers on the local network. This issue has been fixed in version 0.0.28.

A server-side request forgery vulnerability [CWE-918] in Fortinet FortiClientEMS version 7.4.0 through 7.4.2 and before 7.2.6 may allow an authenticated attacker to perform internal requests via crafted HTTP or HTTPS requests.

Apache Allura: sensitive information exposure via DNS rebinding

Import functionality is vulnerable to DNS rebinding attacks between verification and processing of the URL.  Project administrators can run these imports, which could cause Allura to read from internal services and expose them. This issue affects Apache Allura from 1.0.1 through 1.16.0. Users are recommended to upgrade to version 1.17.0, which fixes the issue. If you are unable to upgrade, set "disable_entry_points.allura.importers = forge-tracker, forge-discussion" in your .ini config file.