musickit

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides search functionality that incorporates user-provided text into queries.
  • Ingestion points: The term parameter in the searchCatalog function (SKILL.md) and the searchText state in the MusicSearchView component (references/musickit-patterns.md).
  • Boundary markers: The search string is passed directly to the MusicCatalogSearchRequest without boundary markers or instructions to ignore embedded commands.
  • Capability inventory: The skill snippets are limited to MusicKit API calls and UI updates; they do not include dangerous capabilities such as arbitrary command execution, file system modifications, or non-API network requests.
  • Sanitization: The input is used directly without sanitization or validation.
  • [EXTERNAL_DOWNLOADS]: The skill references an external documentation source for framework information.
  • Source: https://sosumi.ai/documentation/ (Multiple links in SKILL.md)
  • Context: The skill provides links to an unofficial or mirror documentation site for developer guidance on MusicKit APIs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:42 PM
Security Audit — agent-trust-hub — musickit