fusion-360

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a PowerShell command to recursively delete Autodesk Fusion 360 cache directories using the -Force flag. This is a destructive system operation, although documented for troubleshooting sync issues.
  • Evidence: Get-ChildItem -Path "$env:LOCALAPPDATA\Autodesk\Autodesk Fusion 360\*\W.Cache" -Recurse | Remove-Item -Force in gemini.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external CAD models and screenshots, which serves as an entry point for untrusted data. When combined with the skill's ability to write to the local file system, it creates an attack surface for indirect instructions.
  • Ingestion points: Active CAD designs and simulation screenshots (mentioned in gemini.md).
  • Boundary markers: Absent.
  • Capability inventory: Local file system creation and writing (e.g., os.makedirs, exportManager.execute) in SKILL.md and gemini.md.
  • Sanitization: None identified for visual or geometric data processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 03:42 PM