giggle-generation-music

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFEPROMPT_INJECTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests user input for music generation, creating an indirect prompt injection surface.
  • Ingestion points: User descriptions and lyrics provided as input to the prompt parameter in SKILL.md and SKILL.zh-CN.md.
  • Boundary markers: Absent; user text is passed directly to the execution script without delimiters.
  • Capability inventory: The skill can make HTTP POST/GET requests to an external API (giggle.pro) via scripts/giggle_music_api.py.
  • Sanitization: Parameters are handled by the argparse Python library, which treats them as string arguments.
  • [CREDENTIALS_UNSAFE]: The skill follows security best practices by requiring the GIGGLE_API_KEY to be passed via environment variables instead of being hardcoded or passed in command-line arguments.
  • [EXTERNAL_DOWNLOADS]: Network activity is restricted to the legitimate vendor API domain (giggle.pro) using the well-known requests library.
  • [SAFE]: No malicious scripts, obfuscation, or unauthorized data access were found during the analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 12:50 PM