AI & LLM Vulnerabilities

CVE-2025-6855 is a critical vulnerability in Langchain-Chatchat (a tool built on LLMs) up to version 0.3.1 that allows path traversal (accessing files outside the intended directory) through manipulation of a parameter called 'flag' in the /v1/file endpoint. The vulnerability has been publicly disclosed and could potentially be exploited.

CVE-2025-6854 is a path traversal vulnerability (a flaw that lets attackers access files outside intended directories) in Langchain-Chatchat software versions up to 0.3.1, specifically in a file upload endpoint. The vulnerability can be exploited remotely by attackers with login credentials and has already been publicly disclosed.

CVE-2025-6853 is a critical vulnerability in Langchain-Chatchat version 0.3.1 and earlier that allows attackers to exploit a path traversal (a type of attack where an attacker manipulates file paths to access files outside their intended directory) flaw in the upload_temp_docs backend function by manipulating the flag argument. The vulnerability can be exploited remotely by users with basic access permissions, and the exploit details have been publicly disclosed.

Roo Code, an AI agent that writes code automatically, had a vulnerability (CVE-2025-53097) in versions before 3.20.3 where its file search tool ignored settings that should have blocked it from reading files outside the VS Code workspace (the folder a user is working in). An attacker could use prompt injection (tricking the AI by hiding instructions in its input) to make the agent read sensitive files and send that information over the network without user permission, though this attack required the attacker to already control what prompts the agent receives.

iOS Simulator MCP Server (ios-simulator-mcp) versions before 1.3.3 have a command injection vulnerability (a security flaw where attackers insert shell commands into input that gets executed). The vulnerability exists because the `ui_tap` tool uses Node.js's `exec` function unsafely, allowing an attacker to trick an LLM through prompt injection (feeding hidden instructions to an AI to make it behave differently) to pass shell metacharacters like `;` or `&&` in parameters, which can execute unintended commands on the server's computer.

MLflow versions before 3.1.0 have a vulnerability in the gateway_proxy_handler component where it fails to properly validate the gateway_path parameter, potentially allowing SSRF (server-side request forgery, where an attacker tricks the server into making unwanted requests to internal systems). This validation gap could be exploited to access resources the attacker shouldn't be able to reach.

FastGPT, an AI Agent building platform, has a vulnerability in versions before 4.9.12 where the LastRoute parameter on the login page is not properly validated or cleaned of malicious code. This allows attackers to perform open redirect (sending users to attacker-controlled websites) or DOM-based XSS (injecting malicious JavaScript that runs in the user's browser).

Cursor, a code editor designed for AI-assisted programming, had a security flaw in versions before 0.51.0 where JSON files could automatically trigger web requests without user approval. An attacker could exploit this, especially after a prompt injection attack (tricking the AI with hidden instructions in its input), to make the AI agent send data to a malicious website.

FastGPT is an open-source platform for building AI workflows and chatbots that uses a sandbox (an isolated container designed to safely run untrusted code). Versions before 4.9.11 had weak isolation that allowed attackers to escape the sandbox by using overly permissive syscalls (system calls, which are requests programs make to the operating system), letting them read files, modify files, and bypass security restrictions. The vulnerability is fixed in version 4.9.11 by limiting which system calls are allowed to a safer set.

vLLM (a system for running and serving large language models) versions 0.8.0 through 0.9.0 have a vulnerability where the /v1/chat/completions API endpoint doesn't properly check user input in the 'pattern' and 'type' fields when the tools feature is used, allowing a single malformed request to crash the inference worker (the part that actually runs the model) until someone restarts it.

CVE-2025-48943 is a Denial of Service vulnerability (a type of attack that crashes a system) in vLLM versions 0.8.0 through 0.8.x that causes the server to crash when given an invalid regex (a pattern used to match text). This happens specifically when using the structured output feature, which lets the AI format responses in a specific way.