npm detections from LPM Firewall.

The automatic preinstall hook exports an environment credential to a remote service, while validation failures do not enforce a license. That is concrete credential-exfiltration behavior...

The automatic global lifecycle chain modifies foreign AI-agent configurations and injects instructions that can cause conversation and attachment transfer. This meets the install-hook abu...

The automatic lifecycle chain executes an opaque native installer that modifies agent integrations, and the bundled binary contains an unsafe Codex bypass command. This satisfies install-...

The package automatically transmits a named environment credential during preinstall while suppressing invalid-license failures. That concrete credential-exfiltration behavior warrants bl...

This icon package has an automatic installation hook that exfiltrates an environment credential to a remote endpoint. The behavior is concrete and does not require importing the package.

The package contains an automatic preinstall credential-transmission path. Its icon exports do not require this install-time secret disclosure, making the behavior concrete credential exf...

The package performs unconsented install-time credential transmission. Its icon exports do not require this behavior, so the lifecycle hook forms a concrete credential-exfiltration attack...

The package has a concrete install-time credential-exfiltration path. Its icon exports do not require this behavior, and the validation result is non-enforcing.

The automatic lifecycle hook exports an environment credential to a remote server. This is concrete install-time credential exfiltration, so the package should be blocked.

This icon package performs automatic preinstall credential disclosure. The hook transparently targets one named variable, but automatic export of an environment secret to a remote service...

This package automatically transmits an environment credential during preinstall. The source establishes a concrete credential-exfiltration path, so it should be blocked.

This is concrete install-time mutation of a foreign, broad AI-agent control surface. The automatic lifecycle path establishes persistent hooks and global instructions, so it meets the pub...

The source establishes an unconsented postinstall chain that installs a foreign global AI-agent package. The opt-out and detection guard do not change that the mutation occurs automatical...

The package performs an automatic, obfuscated initializer during postinstall that installs its skills into a user-level agent environment. This is concrete install-hook abuse affecting a...

This package performs unconsented postinstall execution of an opaque native agent-hook installer and declares modifications to Codex and Claude Code configuration. That meets the install-...

This is an unconsented postinstall mutation of a foreign AI-agent control surface. The automatic global installation is concrete and sufficient for a blocking verdict under the install-co...

This is a concrete automatic postinstall mutation of multiple foreign AI-agent control surfaces, including default installation and activation of a Codex plugin. The absence of confirmed...

The postinstall hook performs automatic, destructive replacement across three foreign AI-agent skill roots. Opt-out support and a narrow subdirectory name do not establish prior user cons...

This is concrete identity and usage exfiltration to an undeclared hard-coded HTTP endpoint, coupled with persistent agent hook installation. The absence of an npm lifecycle hook does not...

This is unconsented postinstall mutation of broad, foreign AI-agent control surfaces. The opt-out variables and collision checks do not make the automatic behavior consensual.

This package uses postinstall to make unconsented, persistent changes to consumer AI-agent configuration and instructions, including an npx-backed MCP command. That is concrete install-ho...