There has been a significant uptick in attacks targeting Langflow, leveraging an authentication bypass vulnerability that allows unauthenticated remote attackers to fully compromise affected servers, according to a report from FortiGuard Labs.

Langflow is a Python-based web application that offers a visual interface for building AI-driven agents and workflows. An exploit already published for this vulnerability.

A critical authentication bypass vulnerability, identified as CVE-2025-3248, has been discovered in Langflow. This vulnerability allows remote, unauthenticated attackers to execute arbitrary code on affected instances by sending a specially crafted HTTP request with a malicious payload.

CVE-2025-3248 is the case number for the vulnerability, which has a CVSS base score of 9.8.

This vulnerability is easily exploitable and enables unauthenticated remote attackers to fully compromise Langflow servers. There is a patch in Langflow 1.3.0, and users should to upgrade to the latest version.

Schneider Bold

As AI workflows become increasingly prevalent across industries, the need to ensure the security of AI tools and applications has never been more critical. With more reliance on AI systems for decision-making, automation, and innovation, any issue in these systems could have consequences.

Pin It on Pinterest

Share This