npm detections from LPM Firewall.

This is unconsented postinstall mutation of broad, foreign AI-agent control surfaces combined with background persistence. The lack of confirmed remote exfiltration does not remove the co...

The install hook silently writes a persistent workflow into a foreign project and causes future third-party reporting. This concrete install-time mutation is malicious behavior.

The source implements repeated, default-enabled posting of rich runtime records to embedded third-party webhook URLs. This is concrete data exfiltration despite the absence of an install...

The package has a lifecycle-reachable postinstall hook that modifies broad, foreign AI-agent control surfaces without a user-invoked setup action. This meets the install-time AI-agent con...

Although npm installation has no lifecycle hook, the package establishes broad AI-client hooks and remotely replaces the code they execute. This is a concrete remote-code execution and pe...

This is an unconsented postinstall mutation of broad, foreign AI-agent control surfaces. The lifecycle behavior is sufficient for blocking even though the installer itself does not contac...

The opaque curl-to-Bash execution is unrelated to calculating worktime and provides a remote party direct code execution. Its explicit-command trigger reduces exposure but does not remove...

The opaque remote-code execution is hidden behind an unrelated worktime command and gives the short-link destination arbitrary code execution. No install hook is needed for this concrete...

This is unconsented install-time mutation of a foreign AI-agent control surface, with remote shell execution and subsequent automatic agent-message injection. The global-install guard red...

Source inspection confirms automatic install-time system mutation and runtime forwarding of private WhatsApp data to a fixed external recipient. The token gate does not provide consent fr...

The package performs an automatic postinstall global installation of an AI agent CLI and later defaults that agent layer to reduced approval safeguards. This meets the install-hook abuse...

The package performs an automatic postinstall global skill installation into broad AI-agent locations through a downloaded npx tool. This meets the install-hook abuse policy for unconsent...

The package automatically mutates Claude Code's user-level skills directory through an obfuscated postinstall payload. This meets the install-hook policy for an unconsented foreign AI-age...

The package contains an active remote code loader that evaluates unverified server content. Although it has no npm lifecycle hook, this is concrete arbitrary code execution when the bundl...

The automatic lifecycle hook performs unconsented multi-agent configuration mutation and enables a Codex plugin. This meets the install-hook abuse blocking policy despite no confirmed exf...

This is an unconsented postinstall mutation of broad AI-agent control surfaces in the consuming project, combined with an automatic networked self-update path. The source establishes conc...

The source implements a remote-controlled channel for reading arbitrary local dotfiles and returning their contents over a relay, alongside authenticated proxying of a local service. This...

This is a concrete unconsented postinstall mutation of broad AI-agent control surfaces in the consumer project, compounded by changes intended to ensure future lifecycle execution. The ab...

This is an automatic lifecycle-delivered native payload chain that deliberately evades platform download protections and persists through per-user desktop integration. The local hash chec...

The automatic postinstall hook modifies several foreign AI-agent control surfaces and installs activating instructions into each. This is concrete unconsented install-time agent-control m...

The package uses postinstall to mutate a foreign global agent plugin and configuration surface, then installs a plugin that starts detached background execution. This meets the install-ho...