skills/shinyorg/skills/shiny-music/Gen Agent Trust Hub

shiny-music

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests music metadata and lyrics from the local device and an external service (LRCLIB), exposing this data to AI models through the Shiny.Music.Extensions.AI package.
  • Ingestion points: Untrusted content enters the agent's context through metadata returned by IMediaLibrary (titles, artist names, albums) and synchronized lyrics retrieved via ILyricsProvider.GetLyricsAsync from the lrclib.net service.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands within the music metadata or lyrics are defined in the skill documentation.
  • Capability inventory: The skill allows the agent to control playback via IMusicPlayer, manage playlists (create, modify, delete), and copy audio files from the library to local app storage using CopyTrackAsync.
  • Sanitization: There is no mention of sanitization or validation of the fetched metadata or lyrics before they are interpolated into the prompt context for the AI tools.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the LRCLIB web service (https://lrclib.net) to fetch lyrics for tracks. It also references NuGet packages and documentation hosted on the vendor's official GitHub repository (https://github.com/shinyorg/music). These network operations are presented as standard, intended features of the library.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 12:36 AM