npm AI-agent policy reports, with evidence.

This is an automatic postinstall mutation of multiple pre-existing project AI-agent control surfaces. Although no exfiltration or network activity was found, the behavior meets the blocki...

The package combines an automatic install hook with persistent PATH shadowing of Claude and Codex, scheduled execution, and automated self-reinstallation. This is a concrete broad AI-agen...

The automatic postinstall hook performs unconsented writes to foreign and broad AI-agent control surfaces. This meets the install-control-surface block policy even though no install-time...

This is concrete unconsented postinstall mutation of multiple foreign AI-agent configuration surfaces, coupled to package-controlled command execution and dependency patching. Safeguards...

This is an unconsented postinstall mutation of broad foreign AI-agent control surfaces. Checksum verification reduces tampering risk for the downloaded binary but does not establish conse...

The package performs an unconsented postinstall global installation of another AI-agent CLI. That concrete lifecycle mutation warrants blocking even though no credential theft was found i...

This is an unconsented postinstall mutation of a consumer project's broad AI-agent control surface. The safeguards against clobbering do not remove the automatic control-surface write.

The package performs automatic postinstall mutation of both Claude Code and Codex global MCP configuration. This meets the install-control-surface blocking policy even though the inspecte...

The package has a concrete automatic postinstall chain that rewrites Claude Code and Codex configuration to execute its MCP server. This meets the install-hook abuse blocking policy regar...

This is an unconsented postinstall mutation of a broad consumer AI-agent control surface. The installed hooks actively inject policy into agent sessions, which meets the blocking policy e...

This is an unconsented postinstall mutation of a broad global AI-agent control surface. Exclusive creation avoids overwrites but does not remove the risk of first-install persistence.

This is a concrete, automatic postinstall path that executes a remote binary and installs skills into detected AI agents without an explicit user command. Hash verification does not make...

The package has a concrete automatic install-time chain that modifies a consumer project and activates multiple AI-agent integrations. This meets the install-hook abuse policy even though...

This is an unconsented postinstall mutation of broad consumer AI-agent configuration and instructions. The explicit hook command does not remove the automatic MCP and guidance changes.

The package has a concrete automatic postinstall chain that modifies multiple AI-agent homes and installs hooks, rather than requiring an explicit setup command. This meets the install-co...

The package performs unconsented postinstall mutation of consumer AI-agent control surfaces and deliberately removes its own npm traces. That meets the install-control-surface blocking po...

This is an automatic postinstall mutation of a broad AI-agent control surface, not an explicit user-command setup. The recurring hooks and global instruction writes establish concrete per...

This is a concrete unconsented postinstall mutation of global Claude Code configuration and instructions. The package's registry update check does not mitigate the lifecycle control-surfa...

The package automatically persists package-controlled instructions into multiple unrelated AI-agent skill directories during installation and runtime. The opt-out does not make this an ex...

This is an unconsented postinstall mutation of broad AI-agent control surfaces. The forced global installation makes the behavior concrete and block-worthy.

The automatic postinstall mutation of ~/.claude installs package-controlled agent instructions and a command into a pre-existing AI-agent environment without explicit consent. This meets...