fusion-360-printing
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates on untrusted external data, creating an attack surface where malicious instructions could be embedded in design requirements.
- Ingestion points: The skill ingests measurements and design intent from external sources such as videos, notes, and photos, as well as existing CAD geometry (SKILL.md).
- Boundary markers: The instructions do not specify the use of delimiters or boundary markers to separate untrusted user data from the agent's internal modeling instructions.
- Capability inventory: The agent is empowered to use high-capability 'Computer Use' tools for GUI automation and file system operations, including saving and exporting native CAD files and meshes.
- Sanitization: No sanitization or validation logic is defined to check user-provided inputs (parameters, notes) for malicious patterns before they are used to drive the GUI automation.
Audit Metadata