manipulation-ik
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains technical documentation and conceptual Python scripts for robotic simulation within NVIDIA Isaac Sim. No malicious instructions, obfuscation, or unauthorized data access were found.
- [SAFE]: External links reference official NVIDIA documentation and established GitHub repositories, which are recognized as trusted services.
- [INDIRECT_PROMPT_INJECTION]: The skill consumes environment variables and simulation data from the USD stage. The resulting actions are constrained to simulation joint control and exporting data to JSON files, which is consistent with the skill's stated purpose and represents a low-risk attack surface.
- Ingestion points: Environment variables ($ISAAC_SIM_DIR, etc.) and simulation prims (robot_prim, start_prim, end_prim) in SKILL.md and robot_poser_example.py.
- Boundary markers: None.
- Capability inventory: Articulation control (set_dof_position_targets) and file writing (export_poses to JSON).
- Sanitization: None.
Audit Metadata