The Agent Skills Directory

[COMMAND_EXECUTION]: The skill orchestrates its workflow by using subprocess modules in Python and Node.js to execute internal scripts. For instance, run_pipeline.py invokes run_jobs.py, and run_chatgpt_fallback.mjs calls finalize_download.py. These executions are used for process separation and pipeline management rather than executing untrusted external commands.
[EXTERNAL_DOWNLOADS]: The skill interacts with official endpoints for Google Gemini (generativelanguage.googleapis.com) and ChatGPT (chatgpt.com). These are recognized as well-known and trusted services for image generation tasks.
[CREDENTIALS_UNSAFE]: The implementation follows security best practices by retrieving the Google API key from environment variables rather than hardcoding sensitive tokens within the source files.
[DATA_EXFILTRATION]: No patterns of unauthorized data exfiltration were detected. Browser automation state is maintained locally in a persistent profile directory within the user's home folder.

chatgpt-images-fallback