Agentjacking Attack Tricks AI Coding Agents Into Running Malicious Code
Summary: Tenet Security researchers described an attack they call Agentjacking, which injects crafted payloads into Sentry error events so AI coding agents like Claude Code and Cursor read them as legitimate diagnostic steps and run attacker-controlled code. The researchers said they found at least 2,388 organizations exposed with valid injectable DSNs and achieved an 85% exploitation success rate against injected errors across widely used AI coding assistants in controlled testing of over 100 organizations.
Bottom line: Sentry acknowledged the issue but declined to fix it as "technically not defensible," activating only a global content filter that blocks a specific payload string - leaving the architectural trust gap between error ingestion and the Sentry MCP server in place for the 2,388 organizations with injectable DSNs.
Read the original at The Hacker News ↗
- Topics
- AI Security
- Related
- SECURITY AFFAIRS AI-CYBERSECURITY NEWSLETTER ROUND 8 Cyber Risks · Nov 15, 2026Pentagon Renames CDAO ‘Super Intelligence Office’ Cyber Risks · Nov 15, 2026186 Fake AI Domains Redirect Users to Click Fix Lures, Browser Malware, and Search Hijackers Cyber Risks · Nov 12, 2026