According to a recent LinkedIn post from OX Security, the company is drawing attention to critical vulnerabilities it reports having identified in two Model Context Protocol (MCP) platforms, Flowise and Upsonic, with CVSS scores of 10.0 and 9.8, respectively. The post links these findings to an MCP supply chain vulnerability report released last week, which appears to detail input sanitization bypass techniques and associated security lessons.
Claim 55% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
The post suggests that both affected platforms used Anthropic’s recommended input sanitization approach but still remained exposed due to a command execution bypass via NPX using the “-c” flag. According to the shared perspective, this allowed arbitrary commands to be executed on host machines despite restrictions on special characters, underscoring a broader risk in relying solely on input sanitization.
As shared in the post, OX Security emphasizes that security engineers should treat input sanitization as insufficient on its own and instead favor sandboxed execution environments for MCP STDIO servers. The suggested architecture aims to contain command execution, protect sensitive data, and limit lateral movement, which may resonate with enterprises adopting MCP-based tooling.
For investors, the emphasis on novel MCP supply chain vulnerabilities could position OX Security as an early mover in securing emerging AI and automation infrastructure. If the underlying research gains traction with security teams, it may enhance the company’s credibility, support demand for its offerings in the secure software supply chain segment, and potentially strengthen its competitive standing in the cybersecurity market.