I just pushed a change to the OWASP Agentic AI Security Cheat sheet about exactly this, but only as a risk. Now I have an example.

https://www.infosecurity-magazine.com/news/contextcrush-ai-development-tools/

#aiconsole #vulnerability

ContextCrush Flaw Exposes AI Development Tools to Attacks

Critical flaw "ContextCrush" in Context7 MCP Server could allow malicious instructions into AI tools

Infosecurity Magazine