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 11 Oct 2026, 17:33 UTC. Ordered by latest scan.
The automatic lifecycle hook performs unconsented mutation of both Codex and Claude user skill directories, with recursive deletion, and changes the Python user environment. This meets th...
This is an unconsented postinstall mutation of a broad AI-agent control surface, which the policy classifies as malicious install-hook abuse. The absence of a remote payload does not remo...
This is an unconsented install-time remote-payload chain that automatically invokes opaque AI-tool onboarding. The optional integrity check and process termination increase the risk.
This is an automatic lifecycle delivery chain for opaque native code from an unverified remote source, with TLS verification explicitly disabled. It creates a concrete install-time arbitr...