classical-887
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs legitimate network operations to well-known and trusted services, specifically the NPR Composer API and the Spotify Web API. Credential management for Spotify integration follows security standards by using environment variables (SPOTIFY_CLIENT_ID and SPOTIFY_CLIENT_SECRET) or command-line flags rather than hardcoding secrets. File access is limited to the skill's own operational logs and user-directed report generation.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes track metadata (composer names, track titles, and performer details) from the external NPR Composer API. This data is ingested for display and reporting purposes and is not interpolated into executable commands or agent instructions in a manner that could facilitate prompt injection attacks.\n
- Ingestion points:
scripts/classical_887.pyviarequests.getcalls to NPR API endpoints in functionsget_now_playing,get_recent_tracks, andget_tracks_since.\n - Boundary markers: Data is not isolated with specific security delimiters as it is intended for human-readable console and Markdown output.\n
- Capability inventory: Filesystem write access for generating Markdown reports and JSON logs; network access for fetching radio data and managing Spotify playlists.\n
- Sanitization: Employs standard string truncation and JSON serialization for output formatting.
Audit Metadata