Critical Flaw in Flowise AI Tool

A near-maximum severity remote code execution vulnerability affects self-hosted Flowise AI, with reports that the official patch is easily bypassed.

Security researchers at Obsidian Security have uncovered a critical remote code execution (RCE) vulnerability in Flowise, a popular open-source platform for building AI applications. The flaw, rated as near-maximum severity, affects self-hosted deployments of the tool. The vulnerability stems from an issue in how Flowise implements its Model Context Protocol (MCP) stdio servers, which could allow an attacker to execute arbitrary commands on the server with a single click. This type of flaw is particularly dangerous as it can give an attacker significant control over an affected system, potentially exposing sensitive data and internal networks.

What makes this situation especially urgent is the report that the official patch released to fix the vulnerability can be trivially bypassed. This means that even organizations that have applied the update may still be at risk. The vulnerability impacts the many enterprises and developers relying on Flowise for their self-hosted AI workloads. The discovery highlights the ongoing security challenges in the rapidly evolving AI ecosystem and underscores the need for rigorous security auditing of open-source tools, particularly those integrated into enterprise environments.