The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from Spotify (such as song titles, artist names, and podcast descriptions) which could contain adversarial instructions intended to influence the agent's behavior.
Ingestion points: Untrusted data enters the agent context through the spotify-api search, spotify-api experience, and spotify-api library commands mentioned in SKILL.md.
Boundary markers: The instructions do not define explicit delimiters (like XML tags or block quotes) to isolate external content from the agent's system instructions.
Capability inventory: The skill possesses significant capabilities, including modifying the user's music library, controlling device playback, and deleting content via the save-to-spotify CLI as defined in manifest.yaml and SKILL.md.
Sanitization: There are no instructions for sanitizing or filtering external content before the agent interprets or relays it to the user.

spotify