The Agent Skills Directory

[SAFE]: The skill implements secure credential management by instructing the agent to use environment variables (e.g., OPENAI_API_KEY, GOOGLE_GENAI_API_KEY) and encrypted placeholders, rather than hardcoding secrets.
[SAFE]: The script scripts/extract_transparency.py performs purely mathematical image processing using standard libraries (numpy and Pillow). It does not exhibit dangerous behaviors such as subprocess spawning, network exfiltration, or unauthorized file system access.
[SAFE]: External communication is restricted to well-known, legitimate service providers (api.openai.com, googleapis.com, aliyuncs.com).

media-creation