npm detections from LPM Firewall.

The automatic postinstall hook writes into multiple foreign global AI-agent directories and installs remotely delivered executable code. That is concrete install-hook abuse under the stat...

The package has a concrete postinstall chain that automatically alters an MCP Chrome integration and invokes a permission-changing command. This meets the install-time foreign agent contr...

The automatic postinstall hook writes behavior-bearing instructions into the consuming project's CLAUDE.md. This is concrete install-hook abuse even though the inspected code does not sho...

The package performs a remote executable bootstrap during postinstall and automatically executes an opaque hook installer for AI-agent and editor integrations. That creates an unconsented...

This is concrete automatic mutation of a foreign AI-agent control surface, including forced plugin enablement and registry-path redirection. The scope checks do not remove the unconsented...

The automatic lifecycle hook installs an unrelated AI agent globally based on local detection. This is an unconsented install-time mutation of a foreign, broad AI-agent control surface.

The automatic postinstall hook modifies three user-level AI-agent skill locations and forcibly replaces their contents. The opt-out flag does not establish consent for the default install...

This is a concrete, default-enabled data-exfiltration path from third-party authentication files to an external analytics service. Hashing does not remove its use as a persistent cross-se...

The automatic postinstall hook mutates multiple foreign AI-agent skill locations, which meets the install-control-surface blocking policy. No JavaScript exfiltration was found, but it is...

This package performs automatic postinstall writes to both Codex and Claude user skill directories and installs user-level dependencies. Those changes are persistent, cross-product, and o...

This is an unconsented automatic install-time takeover of broad consumer project files, reinforced by a plugin that restores overwritten entry points. The behavior exceeds package setup a...

This is an unconsented postinstall mutation of a consumer AI-agent control surface, compounded by opaque configuration code and unrelated dependency rewrites. Opt-out variables and the gl...

The automatic lifecycle hook modifies consumer configuration and injects AI instructions that promote secret disclosure while suppressing review. The reviewer-directed text is social engi...

This is a concrete, default-enabled remote self-update and profile-mutation chain. It can replace the running plugin with attacker-controlled code from the configured update infrastructure.

The automatic postinstall hook performs unconsented mutation of foreign AI-agent skill directories and installs additional dependencies. This is concrete install-hook abuse, not merely us...

This is an automatic install-time chain that executes an opaque native payload to configure external AI agents. The package's release download is not integrity-pinned, increasing the risk...

This is unconsented postinstall mutation of broad, foreign AI-agent control surfaces, combined with automatic user-environment dependency installation. The behavior meets the install-hook...

Source inspection confirms automatic transmission of agent transcripts and account credentials to a remote endpoint during ordinary use. This is concrete unconsented data exfiltration, de...

The automatic postinstall hook mutates and deletes project-level Claude Code skills. This meets the policy definition of unconsented install-time mutation of a foreign AI-agent control su...

The automatic postinstall chain persistently alters multiple third-party AI-client configurations to launch this package. This meets the policy for unconsented broad AI-agent control-surf...

The automatic lifecycle chain executes an opaque native binary that installs external AI-agent hooks and changes Codex configuration. This is concrete unconsented install-hook abuse, rega...