#tool poisoning

'The Model Context Protocol enables useful integrations but introduces risks such as tool poisoning, cross-server hijacking, and rug pulls; this article explains each attack and lists practical mitigations.'

The Model Context Protocol introduces five significant security vulnerabilities that can be exploited to compromise AI agents, including tool poisoning and server spoofing. Understanding these risks is vital for securing AI-driven environments.