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 13 Oct 2026, 18:06 UTC. Ordered by latest scan.
The package contains no confirmed exfiltration chain, but its automatic lifecycle hook installs executable agent instructions across unrelated AI-agent platforms. That meets the firewall...
The package has no confirmed exfiltration or remote execution, but its postinstall hook concretely and unconsentedly installs active instructions into broad third-party AI-agent control s...
Source inspection confirms the policy-blocking behavior directly in the lifecycle script. CI and opt-out guards do not make the default install-time Claude Code mutation consensual.
The shipped content shows no exfiltration or remote payload behavior, but the automatic lifecycle hook performs concrete, unconsented writes to broad third-party AI-agent instruction surf...