geometry-nodes-for-ovrtx

Warn

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The instructions direct the agent to generate and execute Python code using the blender-python-execution tool and run the local scripts/audit_geometry_nodes.py script within the Blender environment.
  • [REMOTE_CODE_EXECUTION]: The skill instructs the agent to use blender-community-skill-bootstrap to 'obtain' the blender-modeling helper library if it is not already installed. This constitutes a reference to an external and unverifiable source for executable code.
  • [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection via the scripts/audit_geometry_nodes.py script. -- Ingestion points: The script reads object names from the GN_AUDIT_REQUEST global variable or command-line arguments. -- Boundary markers: No delimiters or instructions are used to separate user data from instructions. -- Capability inventory: The script operates within the Blender bpy API, providing access to evaluation and scene data. -- Sanitization: Validation is limited to checking if objects exist and are of the 'MESH' type.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 22, 2026, 06:06 PM
Security Audit — agent-trust-hub — geometry-nodes-for-ovrtx