openusd-stage
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as stage files, prim paths, and metadata documentation strings, which constitutes a potential attack surface.
- Ingestion points: Input parameters for stage files, asset paths, and documentation metadata defined in tools.yaml.
- Boundary markers: No explicit prompt delimiters or instruction-ignoring markers are defined in the script wrappers.
- Capability inventory: The skill performs file system modifications through the dcc_mcp_openusd runtime library.
- Sanitization: Input validation and data handling are delegated to the underlying vendor library implementation.
Audit Metadata