Security audit
Security checks across malware telemetry and agentic risk
This skill mostly matches its content-production purpose, but it needs Review because it combines automatic publishing, stored credentials, background-style material capture, unsafe credential handling, and AI-detection evasion guidance.
Review this skill carefully before installing. Use preview or dry-run modes first, do not provide real publishing credentials until you understand the flows, rotate the hardcoded/exposed API key if it is yours, avoid using it on sensitive conversations because it auto-records material, and remove or change the TLS bypass and anti-detection writing guidance before production use.
data = result.json()
if data["task_status"] == "SUCCEED":
image = Image.open(BytesIO(requests.get(data["output_images"][0]).content))
image.save(output_file)
print(f"图片已保存到: {output_file}")
break67/67 vendors flagged this skill as clean.
Detected: suspicious.dangerous_exec, suspicious.env_credential_access, suspicious.exposed_secret_literal (+2 more)