npm detections from LPM Firewall.

This package automatically mutates global Claude Code configuration and instructions during npm installation, without interactive consent. That is concrete install-hook abuse against a br...

The automatic postinstall chain mutates foreign AI-agent control surfaces and establishes login persistence. This is concrete install-hook abuse under the launcher isolation policy.

The package performs a concrete destructive action automatically at install time against consumer dependencies. This establishes malicious install-hook abuse regardless of the otherwise o...

The package performs automatic postinstall mutation of numerous third-party AI-agent configurations and installs a separate executable MCP dependency. This is a concrete install-hook abus...

The automatic postinstall modification of AGENTS.md and agent skill files is a concrete broad AI-agent control-surface write. Its project and non-clobber guards reduce accidental file los...

This is a concrete automatic postinstall mutation of broad, foreign AI-agent control surfaces. The guarded scope and lack of observed exfiltration do not remove the unconsented persistent...

This is an unconsented install-time chain that fetches and executes opaque remote code and automatically modifies a broad AI-agent control surface. Hash verification does not make the aut...

The automatic lifecycle hook performs broad, destructive mutation of a consumer AI-agent control surface. This meets the install-time control-surface block policy even though no exfiltrat...

This is a concrete automatic postinstall mutation of a broad AI-agent control surface that enables external collection of sensitive interaction content. The opt-out does not establish pri...

The package disguises a large obfuscated command-and-network payload behind an inert middleware export and loads it at import time. Although there is no install hook, the reachable runtim...

The package's automatic postinstall makes privileged host-level software changes and its runtime path repeats the behavior. No exfiltration was confirmed, but the unconsented lifecycle mu...

The install hook automatically performs remote SSH key persistence and starts background services. This is concrete unconsented install-time behavior, not merely a user-invoked SSH utility.

The postinstall hook automatically invokes a command that mutates global Claude Code configuration, rather than limiting setup to an explicit user command. This meets the blocking policy...

Automatic postinstall configuration exports sensitive AI-session content and rewrites global Claude settings without a user command. The accompanying project-wide hook and permission depl...

This is an unconsented postinstall mutation of a broad AI-agent control surface with persistent hook execution. The explicit feedback transport does not offset that install-time behavior.

This package uses automatic install hooks to exfiltrate local environment metadata to an external service. That is concrete unconsented install-time data collection.

The package uses an automatic lifecycle hook to alter global Claude telemetry and enable export of sensitive AI-session content. The opt-out environment variable does not establish prior...

This is malicious install-hook abuse: an automatic postinstall modifies foreign AI-agent control surfaces and installs hooks that collect session context for later telemetry. The user-inv...

The automatic postinstall persists and registers package code in global Claude Code settings for all sessions. This meets the policy for malicious install-hook abuse even without demonstr...

This is a concrete unconsented postinstall mutation of a broad, foreign AI-agent skills directory. Checksum verification and limited telemetry do not remove the risk created by automatic...

This is a concealed import-time remote code execution chain with self-modification, unrelated to a Tailwind forms plugin. The absence of an install hook does not mitigate execution throug...