The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a bundled background removal script locally using a secure method. Evidence: The remove_backgrounds function in scripts/src/gpt_image_cli/cli.py calls subprocess.run with an argument list rather than a shell string, effectively mitigating command injection risks.
[EXTERNAL_DOWNLOADS]: The CLI fetches generated image data from OpenAI's API endpoints. Evidence: scripts/src/gpt_image_cli/cli.py uses urllib.request.urlopen to download image bytes from URLs returned by the official OpenAI SDK. This behavior is consistent with the skill's primary purpose and targets trusted infrastructure.
[SAFE]: The skill implements standard authentication practices by loading API keys from environment files. Evidence: scripts/src/gpt_image_cli/cli.py resolves the OPENAI_API_KEY by checking the current directory and the user's home directory for .env files, which is a common and safe practice for developer tools.

gpt-image