mechanics-visuals
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is designed for local visualization tasks and relies on standard, trusted Python libraries (Matplotlib, NumPy). It does not request network access or sensitive system permissions, and its operations are confined to generating PNG/SVG output based on numerical inputs.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied mechanical parameters to generate diagrams, which serves as a potential surface for indirect prompt injection.
- Ingestion points: User-provided geometry, load values, and sign conventions in SKILL.md.
- Boundary markers: None explicitly present in the instructions to isolate user data.
- Capability inventory: Execution of local scripts (scripts/mech_plots.py) and agent-generated Matplotlib code.
- Sanitization: The provided Python script validates inputs by type-casting numerical parameters to floats, mitigating basic injection attempts.
Audit Metadata