Security checks across malware telemetry and agentic risk
The skill mostly does what it says, but it needs Review because one fal.ai editing path silently disables a provider safety checker and the docs understate that prompts and local images are sent to outside AI services.
Install only if you are comfortable sending prompts and chosen images or masks to OpenAI, Google Gemini, or fal.ai under those providers' terms. Set only the API keys you plan to use, avoid sensitive or regulated images, choose output paths carefully, and review or patch the fal.ai edit path before relying on it because it disables a provider safety checker by default.
if not b64:
url = result.get("data", [{}])[0].get("url")
if url:
with urllib.request.urlopen(url) as img_resp:
return img_resp.read()
raise RuntimeError("No image data in OpenAI edit response")# Try URL fallback
url = body.get("data", [{}])[0].get("url")
if url:
with urllib.request.urlopen(url) as img_resp:
return img_resp.read()
raise RuntimeError("No image data in OpenAI response")66/66 vendors flagged this skill as clean.