Investigate malicious npm packages and AI-agent policy findings. Each report identifies the package version, supporting evidence, advisory sources, and current Firewall decision. Malware findings and policy violations are labelled separately.
Updated 14 Oct 2026, 13:09 UTC. Ordered by latest scan.
The package has a concrete automatic lifecycle chain that modifies a foreign global agent configuration and installs a persistent startup plugin. Source inspection found no need for remot...
The package contains concrete automatic credential and .env exfiltration paths and executes service-supplied shell code. The harmless postinstall banner does not mitigate these runtime be...
The package silently persists itself into Claude Desktop during npm installation, creating automatic execution in a foreign AI-agent control surface. Its subsequent credential access and...
This is a remote-code-execution launcher: an unencrypted, server-controlled binary is automatically updated and executed without verification. Although it has no install hook, the concret...