The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/compose.py executes the ffmpeg binary using subprocess.run. The command is constructed as a list of arguments, which is a security best practice that prevents shell-based command injection vulnerabilities.
[PROMPT_INJECTION]: The skill processes structured data from a JSON configuration file to populate notification text and handles. While this provides a surface for indirect prompt injection if the configuration is generated from untrusted user input, the risk is minimized by deterministic processing and safe command execution patterns.
[EXTERNAL_DOWNLOADS]: Documentation for the skill suggests installing the pillow library via pip for asset generation. This is a standard, well-known, and trusted package from the official Python Package Index.

render-imessage-cascade