AnalysisCybersecurityAI Agents
Jun 12, 12:04 PM
Featured··
Agentjacking Attack Exploits MCP to Hijack AI Coding Agents
Researchers at Tenet Security detail Agentjacking, an attack that uses fake Sentry error events to trick AI coding agents into running attacker-controlled code via MCP. It affects Claude Code, Cursor, and Codex. The attack requires only a public Sentry DSN, no phishing or server compromise.