scenario-minimax-music
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data via the prompt and lyrics parameters to generate music. 1. Ingestion points: prompt and lyrics fields in model_run parameters within SKILL.md. 2. Boundary markers: No explicit delimiters or warnings to ignore instructions are present for these inputs. 3. Capability inventory: The skill utilizes the search, model_run, jobs_wait, and asset_display tools. 4. Sanitization: No sanitization or validation of user-provided lyric strings is mentioned.
- [EXTERNAL_DOWNLOADS]: The skill references the installation of supplementary components from the scenario-labs/skills repository using npx. These are legitimate resources owned by the skill author and represent standard platform extension behavior.
Audit Metadata