claude-code-wingman

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.85). The package intentionally automates and (when configured) auto-approves interactive Claude Code tool prompts (including Bash/tool calls), exposes tmux session output to a monitor/notification daemon that posts prompt details to a webhook/WhatsApp, and provides remote approval controls — together these design choices enable remote arbitrary command execution and facile exfiltration of session content, representing a high-risk backdoor/abuse pattern even if not obfuscated or explicitly exfiltrating credentials by default.