The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection because it processes untrusted user input within its core commands.
Ingestion points: The search, song, and playlists create commands accept arbitrary strings from users or external data sources.
Boundary markers: The skill documentation does not specify the use of delimiters or 'ignore' instructions to separate user input from system logic.
Capability inventory: The skill possesses significant capabilities, including network operations via curl, process execution via node, and local script execution through apple-music.sh.
Sanitization: There is no evidence of input sanitization or validation mechanisms for the parameters passed to the Apple Music API.
[CREDENTIALS_UNSAFE]: The skill requires the management of sensitive authentication material to function.
Evidence: The setup process requires users to provide an Apple Developer Team ID, Key ID, and a .p8 private key file, which are stored in a local config.json file for persistent API access.

apple-music