behavior-tree-generation
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests natural-language scenarios to generate behavior trees, creating a potential surface for indirect prompt injection.\n
- Ingestion points: The
scenariostring passed to thegenerate_behavior_treecoroutine and entered via the Planner panel in the UI.\n - Boundary markers: The provided documentation does not specify the use of delimiters or explicit instructions for the LLM to ignore potentially malicious instructions embedded within the scenario text.\n
- Capability inventory: The resulting behavior trees are used to drive autonomous actor behaviors (characters or robots) through the Isaac Sim Actor SDG system.\n
- Sanitization: No explicit input sanitization or validation logic for the natural-language scenario text is described in the skill's reference materials.
Audit Metadata