cutagent-sound-design
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve processing external sound assets and metadata, which creates a potential surface for indirect prompt injection.
- Ingestion points: The workflow requires reading and auditioning sound effects supplied by the user or from local libraries as specified in references/supplied-assets.md.
- Boundary markers: No explicit instructions are provided to the agent to use specific delimiters or to treat asset filenames/metadata as untrusted content.
- Capability inventory: The skill utilizes the 'CutAgent CLI' and 'CutAgent SDK' to perform project-level operations and file imports in DaVinci Resolve.
- Sanitization: The instructions do not define a specific automated process for sanitizing or validating the contents of asset metadata before they are handled by the agent or the CLI tools.
- [SAFE]: No malicious patterns, unauthorized network activity, or credential exposures were identified. The described interactions with the CutAgent toolset represent intended functionality for the skill's primary purpose.
Audit Metadata