LoLLMs WEBUI provides the Web user interface for Lord of Large Language and Multi modal Systems. A critical Server-Side Request Forgery (SSRF) vulnerability has been identified in all known existing versions of lollms-webui. The @router.post("/api/proxy") endpoint allows unauthenticated attackers to force the server into making arbitrary GET requests. This can be exploited to access internal services, scan local networks, or exfiltrate sensitive cloud metadata (e.g., AWS/GCP IAM tokens). As of time of publication, no known patched versions are available.
parisneo/lollms-webui, in its latest version, is vulnerable to remote code execution due to an insecure dependency on llama-cpp-python version llamacpppython-0.2.61+cpuavx2-cp311-cp311-manylinux231x8664. The vulnerability arises from the application's 'bindingzoo' feature, which allows attackers to upload and interact with a malicious model file hosted on hugging-face, leading to remote code execution. The issue is linked to a known vulnerability in llama-cpp-python, CVE-2024-34359, which has not been patched in lollms-webui as of commit b454f40a. The vulnerability is exploitable through the application's handling of model files in the 'bindingszoo' feature, specifically when processing gguf format model files.
A Cross-site Scripting (XSS) vulnerability exists in the chat functionality of parisneo/lollms-webui in the latest version. This vulnerability allows an attacker to inject malicious scripts via chat messages, which are then executed in the context of the user's browser.
An absolute path traversal vulnerability exists in parisneo/lollms-webui v9.6, specifically in the openfile endpoint of lollmsadvanced.py. The sanitizepath function with allowabsolutepath=True allows an attacker to access arbitrary files and directories on a Windows system. This vulnerability can be exploited to read any file and list arbitrary directories on the affected system.
A Path Traversal and Remote File Inclusion (RFI) vulnerability exists in the parisneo/lollms-webui application, affecting versions v9.7 to the latest. The vulnerability arises from insufficient input validation in the /applysettings function, allowing an attacker to manipulate the discussiondbname parameter to traverse the file system and include arbitrary files. This issue is compounded by the bypass of input filtering in the installbinding, reinstallbinding, and unInstallbinding endpoints, despite the presence of a sanitizepathfromendpoint(data.name) filter. Successful exploitation enables an attacker to upload and execute malicious code on the victim's system, leading to Remote Code Execution (RCE).
A remote code execution (RCE) vulnerability exists in the '/installextension' endpoint of the parisneo/lollms-webui application, specifically within the @router.post("/installextension") route handler. The vulnerability arises due to improper handling of the name parameter in the ExtensionBuilder().buildextension() method, which allows for local file inclusion (LFI) leading to arbitrary code execution. An attacker can exploit this vulnerability by crafting a malicious name parameter that causes the server to load and execute a init.py file from an arbitrary location, such as the upload directory for discussions. This vulnerability affects the latest version of parisneo/lollms-webui and can lead to remote code execution without requiring user interaction, especially when the application is exposed to an external endpoint or operated in headless mode.
A Server-Side Request Forgery (SSRF) vulnerability exists in the 'addwebpage' endpoint of the parisneo/lollms-webui application, affecting the latest version. The vulnerability arises because the application does not adequately validate URLs entered by users, allowing them to input arbitrary URLs, including those that target internal resources such as 'localhost' or '127.0.0.1'. This flaw enables attackers to make unauthorized requests to internal or external systems, potentially leading to access to sensitive data, service disruption, network integrity compromise, business logic manipulation, and abuse of third-party resources. The issue is critical and requires immediate attention to maintain the application's security and integrity.
A path traversal vulnerability was identified in the parisneo/lollms-webui repository, specifically within version 9.6. The vulnerability arises due to improper handling of user-supplied input in the 'listpersonalities' endpoint. By crafting a malicious HTTP request, an attacker can traverse the directory structure and view the contents of any folder, albeit limited to subfolder names only. This issue was demonstrated via a specific HTTP request that manipulated the 'category' parameter to access arbitrary directories. The vulnerability is present in the code located at the 'endpoints/lollmsadvanced.py' file.
A vulnerability in parisneo/lollms-webui versions up to 9.3 allows remote attackers to execute arbitrary code. The vulnerability stems from insufficient protection of the /applysettings and /executecode endpoints. Attackers can bypass protections by setting the host to localhost, enabling code execution, and disabling code validation through the /applysettings endpoint. Subsequently, arbitrary commands can be executed remotely via the /executecode endpoint, exploiting the delay in settings enforcement. This issue was addressed in version 9.5.
A path traversal vulnerability exists in the parisneo/lollms-webui application, specifically within the /listpersonalities endpoint. By manipulating the category parameter, an attacker can traverse the directory structure and list any directory on the system. This issue affects the latest version of the application. The vulnerability is due to improper handling of user-supplied input in the listpersonalities function, where the category parameter can be controlled to specify arbitrary directories for listing. Successful exploitation of this vulnerability could allow an attacker to list all folders in the drive on the system, potentially leading to information disclosure.