1ContextCrush: The Context7 MCP Server Vulnerability Hiding in Plain Sight (opens in new tab)(noma.security)2tcbrah2d ago0
4When Models Examine Themselves: Vocabulary-Activation Correspondence (opens in new tab)(arxiv.org)1tcbrah12d ago0
6Fooling AI Agents: Web-Based Indirect Prompt Injection Observed in the Wild (opens in new tab)(unit42.paloaltonetworks.com)3tcbrah15d ago0