Back to AIBriefs
AnalysisCybersecurityAI Agents

Agentjacking attack tricks AI coding agents into running malicious code

Agentjacking attack tricks AI coding agents into running malicious codeThe Hacker News

Tenet Security researchers describe a new class of attack, Agentjacking, that tricks AI coding agents into executing arbitrary code via fake error reports. A benchmark study also confirms AI coding agents remain vulnerable to prompt injection attacks.

New Attacks Trick OpenClaw AI Agent Into Running Code and Leaking Secrets2 days ago
Frontier AI Models Can Find Crypto's Biggest Bugs. Experts Warn the Industry Isn't Ready5 days agoJason Nelson
Will AI Kill the Bug Bounty Industry?4 days agoKevin Townsend
AI Is Helping Discover Tech Vulnerabilities—And Zcash Is Just the Latest Example5 days agoJason Nelson
Claude Code Vulnerability Could Let Attackers Steal Credentials From GitHub, Says Microsoft5 days agoJason Nelson
AI Agents Still Can't Stop Prompt Injection Attacks, Researchers Warn1 day agoJason Nelson
An active attack is planting backdoors inside Claude Code right now. If you use npm, your credentials may already be compromised.5 days agojohnypitaDiscuss
Control for agentic payments should start at infrastructure4 days agoSignificant-Plant-4Discuss
The Claude Code active attack didn't stop. 294,842 secrets stolen from 6,943 machines. It evolved and now spreads through Python too and uses Claude Code itself to steal your secrets. The risk to your credentials just got bigger.3 days agojohnypitaDiscuss
1 day ago
Agentjacking attack tricks AI coding agents into running malicious code — AIBriefs