cadcraft-floor-plan

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The test suite tests/test_room_plan.py uses importlib.util to dynamically load and execute the local Python module scripts/room_plan.py. This is a standard mechanism for unit testing local components.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined room dimensions and labels to generate CAD command sequences, which represents a potential surface for indirect prompt injection.
  • Ingestion points: User-supplied JSON data read by scripts/room_plan.py from the file specified by the --input argument.
  • Boundary markers: The generated plan does not use explicit boundary markers to isolate user-provided text from the command structure.
  • Capability inventory: The skill is designed to interact with cadcraft-cli and MCP execute tools, which perform file operations and CAD modifications.
  • Sanitization: The build_plan function includes validation for numeric ranges and filters control characters (ASCII < 32) and specific Unicode line separators from the room label.
  • [EXTERNAL_DOWNLOADS]: The skill references a portable binary release of the cadcraft tool on GitHub and suggests the installation of a related skill using npx. These are references to well-known services and related tools from the same author context.
  • [COMMAND_EXECUTION]: The skill provides instructions for executing a local Python script and the cadcraft-cli tool. The documentation includes security best practices, such as using argument arrays instead of shell string concatenation to mitigate command injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 06:49 PM