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 17 Oct 2026, 17:10 UTC. Ordered by latest scan.
Source directly implements covert profile collection and account sweeping, coupled to permission-bypassing AI-agent command execution and platform synchronization. Absence of npm lifecycl...
Direct inspection confirms concrete install-time persistence, remote shell capability, host reconnaissance, and internal credential/service probing. This is malicious behavior unrelated t...
This is a concrete unconsented postinstall mutation of a foreign AI-agent control surface, meeting the blocking policy. The declared vision feature does not remove the lifecycle-control r...
Source directly confirms a postinstall write into ~/.claude/skills and an agent-directed shell-execution payload. This meets the policy boundary for an unconsented lifecycle mutation of a...