Malicious Code Hits MemoryOS and AI-Agent Packages

SlowMist, a security firm, warned developers and crypto infrastructure teams about malicious code inserted into MemoryOS and an AI-agent plugin. The compromised packages include MemoryOS 2.0.34, an open-source library for large language models and AI agents distributed through PyPI. Another affected package is the memtensor/memos-cloud-openclaw-plugin on npm in versions 0.1.21, 0.1.23, and 0.1.25.

The malicious payload is a cross-platform Go binary that executes when an affected package is loaded or imported. It may also expose users' prompt contents. SlowMist recommended removing the affected releases or downgrading to MemoryOS 2.0.33 and npm version 0.1.20, as well as terminating the sckit process and reviewing network activity.

The incident highlights the risk of compromised developer tools being pulled into AI and crypto workflows through routine package installation or updates.