musickit
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted metadata from external search results.
- Ingestion points: Song metadata (titles, artist names) returned by
MusicCatalogSearchRequestinSKILL.mdandreferences/musickit-patterns.md. - Boundary markers: Absent; the instructions do not implement delimiters or warnings to ignore embedded commands within catalog data.
- Capability inventory: The skill has the ability to control playback, manage the playback queue via
ApplicationMusicPlayer, and modify system-level metadata viaMPNowPlayingInfoCenter. - Sanitization: Absent; song titles and artist names are processed and displayed without escaping or validation.
- [METADATA_POISONING]: The
Referencessection inSKILL.mdcontains multiple links to a third-party domain (sosumi.ai) while identifying them as official Apple MusicKit documentation. The use of a non-authoritative domain for framework documentation is deceptive and could mislead users or agents.
Audit Metadata