The Agent Skills Directory

[SAFE]: The skill utilizes a standard authentication workflow via npx -y techdrawai-mcp auth login. This is a routine procedure for Model Context Protocol (MCP) integrations to handle user sessions securely through a browser.
[SAFE]: Credential management instructions recommend using environment variables (TECHDRAWAI_API_KEY), which is a secure and standard practice for preventing hardcoded secrets.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied images as input for CAD conversion.
Ingestion points: The image parameter in the image_to_dxf, photo_to_3d, and render_product tools (SKILL.md).
Boundary markers: None explicitly defined for image content, as images are processed as binary data.
Capability inventory: File writing to a default or user-defined output_path (SKILL.md).
Sanitization: The conversion of raster images into structured vector geometry (DXF) or 3D meshes (STL/GLB) inherently prevents the execution of embedded natural language instructions, as the output is non-textual geometric data.

techdrawai-cad