solidworks-design-vault-cad

Warn

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to clone a repository from a source not recognized as a trusted organization (github.com/fgiafrica/SolidWork-Design-Vault).
  • [COMMAND_EXECUTION]: Includes instructions to execute shell commands for repository cloning and provides Python examples using subprocess.run to invoke local system binaries like swScheduler.exe.
  • [DYNAMIC_EXECUTION]: Provides scripts utilizing powerful automation interfaces, including the Windows COM bridge (win32com.client), SolidWorks VBA macros, and C# code execution within Node.js environments via the edge-js library.
  • [INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface by processing external CAD files while maintaining high-privilege automation capabilities.
  • Ingestion points: The skill is designed to load and process .SLDPRT, .SLDASM, and .SLDDRW files from external sources.
  • Boundary markers: No explicit boundary markers or safety instructions are provided to the agent regarding untrusted CAD metadata.
  • Capability inventory: The skill utilizes subprocess execution, COM automation, and dynamic C# execution.
  • Sanitization: There are no apparent validation or sanitization checks in the provided script templates to handle malformed or malicious CAD file metadata.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 16, 2026, 03:58 AM