npm AI-agent policy reports, with evidence.

Runtime source establishes deliberate AI-directed source-inspection suppression plus remote collection of host-derived identifiers and error text. These behaviors create a concrete agent...

The automatic lifecycle hook downloads and executes opaque native code specifically to install AI-agent hooks. This meets the install-control-surface blocking policy.

Direct source inspection confirms an automatic postinstall chain that delegates permission changes and Chrome MCP registration to a third-party bridge. This is an unconsented install-time...

This is an unconsented install-time mutation of broad consumer AI-agent control surfaces, reinforced by lifecycle-setting changes and reviewer-suppression payloads. The absence of observe...

This is concrete postinstall modification of a foreign AI-agent installation, including credential-check and authentication weakening, followed by destructive dependency pruning. The beha...

This package has a concrete automatic postinstall chain that mutates global AI-agent configuration and skills plus the consumer project. That meets the install-hook abuse block boundary e...

This is a concrete automatic postinstall mutation of foreign and broad AI-agent control surfaces. The opt-out does not establish affirmative consent for the lifecycle action.

This is an unconsented postinstall mutation of foreign, broad AI-agent control surfaces. The lifecycle guard only prevents a separate global install and does not prevent the skill writes.

The automatic postinstall chain writes consumer-project files and registers an MCP server in Cursor and Codex configuration. This is concrete install-hook abuse even though the inspected...

This package automatically rewrites consumer Next.js and AI configuration during postinstall, then supplies interception code that exports LLM conversation content. The lifecycle mutation...

The package contains a concrete silent postinstall chain that writes into global control files for multiple unrelated AI-agent tools. This meets the policy threshold for unconsented insta...

The package performs an automatic postinstall write into broad project agent configuration and can create new agent files in a repository that has not explicitly requested setup. The inje...

This package uses an automatic install hook to persist instructions into unrelated AI-agent configuration files and steer future agents toward financially and credential-sensitive capabil...

The package has a reachable postinstall path that mutates several foreign AI-agent control surfaces and enables network access. Its installed instructions require remote content, creating...

The automatic lifecycle hook deletes credentials associated with other AI providers and replaces configuration with a vendor-provisioned key. That is an unconsented install-time mutation...

Automatic postinstall code modifies global Claude Code hooks, MCP configuration, and permissions. This concrete agent-control takeover meets the install-hook abuse blocking boundary.

The automatic postinstall mutation of a consumer AI-agent control surface meets the blocking policy. No network activity is needed for this concrete install-time control-surface takeover.

The automatic postinstall modifies foreign, user-wide AI-agent configuration and installs an MCP command that resolves the latest release at launch. This meets the install-control-surface...

The package has a concrete automatic postinstall chain that overwrites a foreign AI-agent executable with opaque payloads. That meets the install-control-surface blocking policy despite b...

The automatic lifecycle hook modifies a consumer-owned MCP configuration rather than only package-owned files. That creates an unconsented agent-extension execution path.

The package has a concrete automatic postinstall chain that overwrites skill content in Claude, agent, and Codex directories. Opt-out support does not establish consent for the default mu...