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:
- The security dilemma of smart factories [Part 1] Specificity of the programming languages used to move industrial robots
March 26, 2021
Industrial robots are the core of the automation of manufacturing processes in smart factories, and are the most important components as they support the manufacture of all kinds of products such as automobiles, aircraft, processed foods, and pharmaceuticals. In addition, as equipment that realizes unmanned manufacturing in the post-COVID-19 world where minimal or no contact ...
- Microsoft Exchange servers now targeted by Black Kingdom ransomware
March 22, 2021
Another ransomware operation known as ‘Black Kingdom’ is exploiting the Microsoft Exchange Server ProxyLogon vulnerabilities to encrypt servers. Over the weekend, security researcher Marcus Hutchins, aka MalwareTechBlog, tweeted that a threat actor was compromising Microsoft Exchange servers via the ProxyLogon vulnerabilities to deploy ransomware. Based on the logs from his honeypots, Hutchins states that the threat actor ...
- Hacking group used 11 zero-days to attack Windows, iOS, Android users
March 20, 2021
Project Zero, Google’s zero-day bug-hunting team, discovered a group of hackers that used 11 zero-days in attacks targeting Windows, iOS, and Android users within a single year. The Project Zero team revealed that the hacking group behind these attacks ran two separate campaigns, in February and October 2020. This month’s report showcases the use of seven zero-days ...
- Convuster: macOS adware now in Rust
March 18, 2021
Traditionally, most malicious objects detected on the macOS platform are adware: besides the already familiar Shlayer family, the TOP 10 includes Bnodlero, Cimpli, Adload and Pirrit adware. As a rule, most tend to be written in C, Objective-C or Swift. Recently, however, cybercriminals have been paying increased attention to new programming languages, seemingly in the ...
- Cisco Plugs Security Hole in Small Business Routers
March 17, 2021
A popular line of small business routers made by Cisco Systems are vulnerable to a high-severity vulnerability. If exploited, the flaw could allow a remote – albeit authenticated – attacker to execute code or restart affected devices unexpectedly. Cisco issued fixes on Wednesday for the flaw in its RV132W ADSL2+ Wireless-N VPN routers and RV134W VDSL2 ...
- Critical Security Hole Can Knock Smart Meters Offline
March 12, 2021
Critical security vulnerabilities in Schneider Electric smart meters could allow an attacker a path to remote code execution (RCE), or to reboot the meter causing a denial-of-service (DoS) condition on the device. Schneider Electric’s PowerLogic ION/PM smart meter product line, like other smart meters, is used by consumers in their homes, but also by utility companies ...