bmz-lyrics-check
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted user-provided lyrics to perform detailed analysis and multi-source network searches (e.g., via Google, Baidu, and music platforms).
- Ingestion points: User lyrics provided via text or screenshots.
- Boundary markers: The skill lacks explicit instructions to treat input as data rather than instructions, potentially allowing a user to include malicious directives within the lyric text.
- Capability inventory: Browsing/searching capabilities are used to perform originality checks and copyright verification.
- Sanitization: No explicit sanitization or filtering of external content is defined before it is used in search queries.
- [EXTERNAL_DOWNLOADS]: The skill instructions mandate the use of network search tools to verify lyric originality across various trusted and well-known music and search platforms (Google, YouTube, NetEase Music, QQ Music, etc.). These operations are strictly functional and serve the primary purpose of copyright verification.
- [SAFE]: The skill is authored by 'bmz404' and consistently references vendor-specific methodology and resources. All external URLs point to well-known music services or legitimate search engines.
Audit Metadata