Security audit
Security checks across malware telemetry and agentic risk
The skill mostly matches a PRD workflow, but it also mutates the host during install and ships unrelated internal-login automation with hardcoded credentials, so it needs review before installation.
Review before installing, especially in a work or sensitive environment. Remove or disable the postinstall hook, delete and rotate the hardcoded internal credentials, constrain output paths, replace shell-string and AppleScript path handling with safe argument APIs, and update the documentation to clearly disclose CDN/API use, browser automation, local storage paths, subagent use, and cleanup behavior.
# 自动打开
if args.open:
print(f"\n🌐 打开 HTML...")
os.system(f"open {html_path}")
# 完成总结
print("\n" + "=" * 60)tell application "Safari" to close window 1
'''
result = subprocess.run(
['osascript', '-e', script],
capture_output=True,
text=True,print(f" 命令：{' '.join(cmd)}")
try:
result = subprocess.run(
cmd,
capture_output=True,
text=True,64/64 vendors flagged this skill as clean.
Detected: suspicious.dangerous_exec