This blog details research and analysis of an active campaign that exploits a critical unauthenticated remote code execution (RCE) vulnerability, CVE-2025-3248, that has been identified in Langflow versions prior to 1.3.0. Langflow is a Python-powered visual framework for building AI applications with over 70,000 GitHub stars, and its versions prior to 1.3.0 contains a flaw in its code validation mechanism that permits arbitrary code execution.
Unauthenticated attackers can exploit this vulnerability by crafting malicious POST requests to the /api/v1/validate/code endpoint. The malicious payload in our investigation was found embedded within argument defaults or decorators of a Python function definition. Since Langflow does not enforce input validation or sandboxing, these payloads are compiled and executed within the server’s context, leading to RCE.
Read more…
Source: Trend Micro
Sign up for our Newsletter
The latest news and insights delivered right to your inbox.
Related:
- IBM Patches Reflected XSS in Worklight, MobileFirst
August 2, 2017
BM fixed a cross-site scripting vulnerability in two products last month that could have let an attacker execute malicious JavaScript code in a victim’s browser to steal sensitive information, or user credentials. The vulnerability (CVE-2017-1500) lingered in the products, Worklight and MobileFirst, for almost a year. Gabriele Gristina, a security consultant for the Italian information security ...
- Attack Uses Docker Containers To Hide, Persist, Plant Malware
July 27, 2017
A novel attack vector allows for adversaries to abuse the Docker API to hide malware on targeted systems, and even execute remote code. The proof of concept attack was developed by researchers at Aqua Security, and the technique was first demonstrated today at Black Hat by Sagie Dulce, senior security researcher, with Aqua Security. The attack works ...
- CowerSnail — Windows Backdoor from the Creators of SambaCry Linux Malware
July 27, 2017
Last month, we reported about a group of hackers exploiting SambaCry—a 7-year-old critical remote code execution vulnerability in Samba networking software—to hack Linux computers and install malware to mine cryptocurrencies. The same group of hackers is now targeting Windows machines with a new backdoor, which is a QT-based re-compiled version of the same malware used to ...
- Novel Attack Tricks Servers to Cache, Expose Personal Data
July 25, 2017
Researcher Omer Gil has devised a way to trick a web server into caching pages and exposing personal data. The so-called web caching attack targets sites that use content delivery network (CDN) services such as Akamai and Cloudflare. These services act as traffic load balancers and reverse proxies, and store files that are frequently retrieved in ...
- Apple Patches BroadPwn Bug in iOS 10.3.3
July 20, 2017
Apple released iOS 10.3.3 Wednesday, which serves as a cumulative update that includes patches for multiple vulnerabilities including the high-profile BroadPwn bug that allowed an attacker to seize control of a targeted iOS device. BroadPwn was revealed earlier this month as a flaw in Broadcom Wi-Fi chipsets used in Apple and Android devices. Apple said the ...
- Critical Code Injection Flaw In Gnome File Manager Leaves Linux Users Open to Hacking
July 19, 2017
A security researcher has discovered a code injection vulnerability in the thumbnail handler component of GNOME Files file manager that could allow hackers to execute malicious code on targeted Linux machines. Dubbed Bad Taste, the vulnerability (CVE-2017-11421) was discovered by German researcher Nils Dagsson Moskopp, who also released proof-of-concept code on his blog to demonstrate the ...