music-prompt
Warn
Audited by Socket on Sep 16, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The core prompt-writing behavior is benign and aligned to the stated purpose, and the command-injection findings are false positives from documentation markup. Risk comes from the optional execute layer: it expands a text-only skill into an API-calling runner, routes Suno through a user-specified gateway because no official API exists, and forwards credentials/prompts to third-party routers like Replicate and fal.ai. Those data flows are plausible for this domain but materially increase trust and credential-handling risk, especially since install.sh and dependency contents were not provided.
Confidence: 89%Severity: 74%
Audit Metadata