New 'Agentjacking' Attack Uses Fake Sentry Errors to Hijack AI Coding Agents

Researchers at Tenet Security (Tenet Threat Labs) disclosed in June 2026 a new attack class called "Agentjacking," which uses fake Sentry error reports to hijack AI coding agents such as Claude Code and Cursor and run attacker-controlled code on a developer's machine. No phishing, malware or compromise of a target server is required — the attack starts with nothing more than a single fake error event.