solidworks-fillet-chamfer-cnc

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined geometric parameters to generate SolidWorks models, which establishes an ingestion point for data that could theoretically influence the agent's behavior if the data is processed by reviewing tools.
  • Ingestion points: User-provided parameters in scripts/create_cnc_mount_template.py via the MountParameters dataclass (e.g., length, width, thickness).
  • Boundary markers: Absent. The skill does not employ specific delimiters or "ignore" instructions for the ingested parameter data.
  • Capability inventory: The skill possesses file-writing capabilities (SLDPRT, STEP, and JSON formats) and executes complex operations through the SolidWorks COM API.
  • Sanitization: Absent. While parameters are cast to floating-point numbers, there is no semantic validation or sanitization to prevent logic-based injection targeting the CAD engine or analysis scripts.
  • [COMMAND_EXECUTION]: The skill relies on the execution of local scripts and API interactions to automate the CAD environment.
  • Evidence: SKILL.md instructs the agent to execute python ../../scripts/sw_preflight.py and sw_review.run_review(), which involve process management and file system operations on the host machine.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 09:53 AM
Security Audit — agent-trust-hub — solidworks-fillet-chamfer-cnc