deduplicate-hierarchies

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute the usdOptimize command-line utility for USD hierarchy deduplication. This is a legitimate functional requirement for asset optimization within the Omniverse ecosystem.
  • [PROMPT_INJECTION]: The skill instructions allow for user-provided file paths and prim roots to be used as arguments in shell commands, which is a common surface for indirect prompt injection.
  • Ingestion points: The <asset> path and --paths input arguments defined in the SKILL.md inputs section.
  • Boundary markers: The skill does not implement specific markers or instructions to isolate user-provided strings from the shell command context.
  • Capability inventory: Shell command execution via the Bash tool (Step 3).
  • Sanitization: The instructions do not specify input validation or shell escaping for the provided paths, relying on standard agent safety protocols.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 07:55 AM
Security Audit — agent-trust-hub — deduplicate-hierarchies