discover-music

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill directly fetches and ingests open/public third‑party content — e.g., ListenBrainz API calls in src/listenbrainz_client.py, MusicBrainz calls in src/musicbrainz_client.py, Bandcamp scraping noted in docs/01_TASKS.md, and YouTube transcript/download delegation in src/cli_youtube.py — and that external/user-generated content is parsed and used to drive recommendations, reranking, queuing for training, and downloads, so untrusted content can materially influence agent behavior.