malware
Malicious File in Downloads Folder Tries to Trick AI Coding Agent Into Installing Malware
Medium
Summary
A security firm, Forever Security, discovered a malicious file-dropping technique targeting AI coding agents like OpenAI's Codex and Anthropic's Claude. OpenAI and Anthropic are not patching this issue, as the agent is designed to trust instructions files. The firm advises users to avoid running agents in their Downloads folder and to carefully examine agent configuration files for unauthorized changes.
Summary written automatically in our own words from the original article, which belongs to its publisher and remains the reference. It may contain errors. Sources & data