blender-editable-3d
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions designed to influence the agent's code generation process.
- Ingestion points: Processes user-provided reference images, design drawings, and text concepts as described in
SKILL.md. - Boundary markers: The instructions do not specify explicit delimiters or "ignore embedded instructions" warnings for the ingested data.
- Capability inventory: The skill has the capability to generate and execute Blender Python scripts and write multiple file formats (.blend, GLB, PNG) to the local filesystem.
- Sanitization: No specific sanitization or validation mechanisms for the generated code are mentioned to prevent command injection or unauthorized API calls.
- [DYNAMIC_EXECUTION]: The core functionality of the skill involves the automated generation and execution of scripts, which is a common but sensitive operation.
- Evidence:
SKILL.mdinstructs the agent to "understand references and write Blender Python" and execute these scripts using the Blender CLI (blender --python). This represents dynamic script generation and execution based on AI-interpreted requirements.
Audit Metadata