very-happy-cli@0.2.118
Remote multi-agent workspace CLI with deep Claude integration, native terminal streaming, and self-hosting.
AI security review
scanned · by lpm-firewall-aiThis is not an install-hook attack: postinstall only prepares local tools, while relay and Claude-hook behavior require user commands. It remains a dangerous remote AI-agent capability because the configured relay can access session content and act through an online daemon.