@funny-booth/agent-core@0.1.68
AIエージェント起動CLI(共有ランチャー + MCPサーバー)
AI security review
scanned · by lpm-firewall-aiThe package contains an agent instruction that prompts feedback submission without user approval, and the feedback function transmits operational details to a portal. The instruction limits reported content to tool-suite issues and excludes project data and secrets, but this is still undisclosed runtime telemetry.