The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection by ingesting natural language instructions to drive its autonomous 3D modeling tools.
Ingestion points: workflow_prompt in agentic_blender_workflow and prompt in intelligent_3d_processing.
Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands embedded within user-provided modeling descriptions.
Capability inventory: The skill has access to tools that can create, modify, and texture objects within a 3D environment, potentially affecting scene integrity if subverted.
Sanitization: No sanitization or input validation logic is described for the prompts processed by the workflow tools.
[DYNAMIC_EXECUTION]: The skill documentation highlights "script-based creation" and uses an agentic workflow tool to execute multi-step modeling tasks.
Evidence: The agentic_blender_workflow tool is designed to plan and execute modeling operations autonomously from natural language, which involves the dynamic generation and execution of command sequences or scripts within the Blender API context.

autonomous-modeling