npm AI-agent policy reports, with evidence.

Direct source inspection confirms a concrete AI-agent control-surface mutation that exfiltrates agent transcripts to a hardcoded remote endpoint without clear user-facing disclosure in RE...

The package performs unconsented lifecycle writes to OpenCode agent/control files with shell/edit-capable agents, matching the firewall's malicious AI-agent control hijack boundary. Other...

Source inspection confirms unconsented lifecycle mutation of AI-agent control files, which is a concrete firewall block condition independent of whether the wallet features are otherwise...

Static inspection confirms concrete install-time AI-agent control-surface mutation, which matches the firewall block boundary even if the runtime wallet/network code is otherwise package-...

Direct source inspection confirms unconsented lifecycle writes into AI-agent control-surface directories from remote content, which meets the firewall blocking criterion for AI-agent cont...

The confirmed lifecycle write to ~/.claude/skills is an unconsented AI-agent control-surface mutation, which is blockable even without credential theft or external exfiltration. Other ris...

This is malicious under the firewall boundary because lifecycle install code writes package-supplied AI-agent instructions into a global Claude skills directory. The behavior is not limit...

Static source inspection confirms an install-time lifecycle script writes package-supplied Claude skill instructions into the user's global AI-agent configuration. Under the firewall boun...