The Agent Skills Directory

[SAFE]: No malicious code or suspicious activity was identified. The skill follows best practices for Blender scripting and Sverchok node-based modeling.
[PROMPT_INJECTION]: The skill provides templates for ingesting external data via CSV and JSON files, which represents an attack surface for indirect prompt injection if untrusted data is processed.
Ingestion points: SvCSVInNode and SvJsonInNode are used in SKILL.md (Pattern 6), references/examples.md (Examples 9, 10), and references/methods.md to load external spreadsheets and architectural data.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the data processing examples.
Capability inventory: The skill's capabilities are focused on geometry generation and Blender object manipulation via the bpy and sverchok APIs. No dangerous capabilities such as shell execution or network exfiltration were found.
Sanitization: The examples use standard parsing methods like json.loads without additional data validation or sanitization for potential embedded instructions.

sverchok-impl-parametric