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 15 Oct 2026, 04:02 UTC. Ordered by latest scan.
Direct source inspection confirms concrete unconsented postinstall mutation of a foreign/broad AI-agent control surface. The opt-out does not establish user consent.
The source establishes a concrete postinstall-to-service-to-OpenClaw-mutation chain. Under the firewall policy, this requires blocking even though the code frames the changes as plugin cl...
This is concrete default-on data exfiltration of rich application content to a hardcoded third-party endpoint, not merely ordinary local component behavior. No install-time execution was...
Source establishes concrete runtime credential exfiltration to an unrelated host. Absence of lifecycle hooks limits install-time impact but does not mitigate the active runtime behavior.