npm detections from LPM Firewall.

The postinstall hook mutates a foreign, broad AI-agent command surface by claiming codex and claude globally. The later launcher configures an unrestricted agent and redirects it to packa...

The automatic postinstall hook changes Claude Code skills in the consumer project and removes existing skill directories. This meets the install-time AI-agent control-surface abuse policy...

The package performs unconsented postinstall mutation of a foreign AI-agent control surface by importing Claude Code extensions. This is concrete install-hook abuse even though the inspec...

This is concrete unconsented postinstall mutation of broad, foreign AI-agent control surfaces. The cross-platform fan-out and automatic installation meet the install-hook abuse blocking p...

The automatic lifecycle hook broadly changes AI-agent configuration and instruction directories across multiple vendors. This matches install-hook abuse even though no network exfiltratio...

This package automatically overwrites skills across three agent ecosystems during installation. That is concrete install-hook abuse even though the installer itself has no network or cred...

This is an unconsented postinstall execution chain that installs AI-agent hooks through an opaque native binary. The visible source provides no bounded list of affected settings or integr...

This is concrete unconsented postinstall mutation of a foreign AI-agent control surface. The lack of direct secret theft does not neutralize the install-hook abuse.

The package performs unconsented postinstall mutation of existing workspace and agent MCP control surfaces. That is a concrete install-hook abuse path even though the reviewed code did no...

This is unconsented postinstall mutation of broad, foreign AI-agent control surfaces, including a future remote package execution path. The absence of observed secret theft does not remov...

This is unconsented postinstall mutation of a broad, foreign AI-agent control surface. No data theft was confirmed, but the automatic persistent agent-control changes meet the publish-blo...

The package contains a concrete runtime credential redirect to an unrelated fixed endpoint, not merely ordinary API integration. Its main entry exports the feature, making the behavior di...

This is an automatic postinstall path that invokes an opaque native binary to reconcile AI-client hooks and state. That is a concrete unconsented lifecycle mutation of a foreign AI-agent...

This package automatically mutates a foreign AI-agent control surface during postinstall and supplies a launcher that enables unrestricted permissions and changes user agent configuration...

The automatic postinstall hook silently delegates to an opaque native binary to modify several foreign coding-agent skill directories. This meets the install-control-surface block policy.

This is an automatic, obfuscated install hook that downloads and executes unverified native code from a remote host. Its dependency-repair purpose does not remove the install-time remote-...

The package contains a concrete credential collection and transmission path to a hard-coded external endpoint. Although it has no install lifecycle hook, this behavior is sufficient crede...

This is an automatic postinstall mutation of multiple pre-existing project AI-agent control surfaces. Although no exfiltration or network activity was found, the behavior meets the blocki...

Source inspection confirms a complete credential capture and upload path to a remote default endpoint. The broad browser permission and unrelated default host make this malicious credenti...

The package contains concrete, automatic install-time host mutation and network download behavior. Although the implementation is readable and no credential theft was found in the inspect...

The package combines an automatic install hook with persistent PATH shadowing of Claude and Codex, scheduled execution, and automated self-reinstallation. This is a concrete broad AI-agen...