validate-album

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses a local configuration file at ~/.bitwize-music/config.yaml and reads album data from directories defined in the content_root and audio_root. This access is necessary for the skill's stated purpose of album validation and does not involve sending data to external or untrusted domains.
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform structural checks (e.g., test -d, test -f, ls). It also generates mv commands to suggest fixes for misplaced directories. These commands are localized to the user's music library and are consistent with the skill's utility functions.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests and parses the contents of untrusted markdown files within the tracks/ directory.
  • Ingestion points: Reads all track metadata files located at {album_path}/tracks/*.md using the Read tool.
  • Boundary markers: None. The skill processes the content of these files without explicit delimiters or instructions to ignore embedded agent commands.
  • Capability inventory: The agent has access to Bash and the bitwize-music-mcp toolset, which can modify the file system.
  • Sanitization: None. The skill extracts specific fields (Status, Suno Inputs, etc.) but does not sanitize the remaining text for potential instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 10:05 PM