npm AI-agent policy reports, with evidence.

The confirmed postinstall global AI-agent skill installation meets the blocking policy. Runtime daemon helpers do not reduce that install-time risk.

This is a concrete, unconsented postinstall mutation of a foreign/broad AI-agent control surface. The opt-out environment variable does not make the default install behavior consensual.

The postinstall hook performs unconsented mutation of a foreign, broad AI-agent control surface. This meets the install-control-surface block boundary.

Source inspection confirms an unconsented postinstall operation that installs third-party skills globally into ~/.agents/skills/. Opt-out and non-fatal handling do not make the mutation u...

Source inspection confirms an unconsented postinstall write to a foreign AI-agent skills directory through a remote npx command. This meets the install-control-surface block policy.

This is a concrete unconsented lifecycle-chain mutation of persistence and a foreign AI-agent launch surface, not merely dormant CLI capability.

This is a concrete unconsented postinstall mutation of foreign/broad AI-agent control surfaces. The absence of exfiltration does not remove the install-time control-hijack behavior.

Direct source inspection confirms unconsented postinstall mutation of broad foreign AI-agent control surfaces and OS persistence. This meets the install-control-surface policy for blocking.

This is concrete unconsented postinstall mutation of broad foreign AI-agent control surfaces, compounded by download-and-execute of an unverified remote launcher.

Direct inspection confirms a concrete default postinstall control-surface mutation, not merely a scanner heuristic. Opt-out and idempotence do not provide prior consent.

Source directly establishes both remote binary execution and silent mutation of multiple third-party/global AI-agent instruction files during postinstall. This meets the block policy rega...

Source confirms unconsented postinstall mutation of a broad, foreign AI-agent control surface. Lack of network activity does not mitigate this concrete install-time control injection.

This is a concrete, unconsented install-time mutation of broad third-party AI-agent control surfaces. It meets the firewall block policy even though no credential exfiltration was observed.

The concrete install-time mutation of foreign AI-agent discovery directories meets the blocking policy. No separate exfiltration chain is needed for this control-surface hijack.

This is concrete unconsented postinstall mutation of foreign/broad AI-agent control surfaces, meeting the blocking policy. The absence of observed exfiltration does not mitigate that inst...

The package has a concrete automatic postinstall chain that installs its agent instructions into every detected supported platform. Local-only CLI behavior does not mitigate that unconsen...

This is a concrete, unconsented postinstall mutation of broad foreign AI-agent control surfaces. Safeguards against source-code changes and exfiltration do not remove that control-hijack...

Direct source inspection confirms install-time modification of a third-party AI-agent dependency. This meets the firewall block boundary for unconsented postinstall mutation of a foreign...

Source inspection confirms the critical behavior independently of scanner output. The lifecycle hook meets the firewall block policy despite no observed exfiltration or network activity i...

Source confirms unconsented install-time mutation of the broad Codex skill control surface. This meets the firewall block boundary regardless of the otherwise conventional CLI authenticat...

This is a concrete unconsented postinstall mutation of broad, foreign AI-agent control surfaces. It meets the firewall block boundary even though the postinstall itself has no observed ne...