classical-887

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [Data Exposure & Exfiltration] (LOW): Performs network requests to api.composer.nprstations.org and api.spotify.com, which are not on the default whitelist. These connections are necessary for fetching radio data and managing Spotify playlists.
  • [Indirect Prompt Injection] (LOW): The skill processes untrusted metadata from the NPR API and uses it in subsequent actions like Spotify searches and report generation.
  • Ingestion points: NPR Composer API (api.composer.nprstations.org).
  • Boundary markers: No delimiters are specified to isolate external data.
  • Capability inventory: Writing Markdown/JSON files and modifying Spotify playlists.
  • Sanitization: No sanitization methods are mentioned in the documentation.
  • [External Downloads] (SAFE): Relies on standard and trustworthy Python libraries including requests and spotipy.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 05:49 PM