blender-modeling-modifiers
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill constructs Python scripts using names and properties retrieved from the Blender environment, such as active object names. This presents a standard surface for indirect injection if a Blender project contains malicious metadata that could be interpolated into a command.
- Ingestion points: Environment data such as object names used in
SKILL.mdandreferences/python_api.md. - Boundary markers: The skill does not currently implement specific delimiters or instructions to treat environment strings as untrusted data.
- Capability inventory: The
execute_blender_codetool is used to run the resulting Python scripts, granting the ability to perform scene mutations. - Sanitization: No explicit escaping or validation logic is provided for the interpolated strings.
- [DYNAMIC_EXECUTION]: The skill uses template-based Python script generation to be executed by the Blender platform tools. This dynamic assembly is the primary method of operation and is consistent with the skill's intended use-case.
Audit Metadata