matlab-design-pcb-txline

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill contains legitimate technical instructions and code snippets for electronic design automation within MATLAB. All operations use standard toolbox functions (e.g., microstripLine, stripLine, pcb2D) and target engineering workflows without any evidence of malicious activity or unauthorized system access.
  • [PROMPT_INJECTION]: The skill was analyzed for indirect prompt injection risks (Category 8) as it transforms user-specified design requirements into executable MATLAB code.
  • Ingestion points: User-provided input such as trace dimensions, material properties, and operating frequencies defined in prompts (SKILL.md).
  • Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters or protective wrappers when interpolating user data into code blocks.
  • Capability inventory: The skill utilizes evaluate_matlab_code and check_matlab_code for code execution (manifest.yaml).
  • Sanitization: Not explicitly present in the skill instructions; the safety of the generated code relies on the agent's internal guardrails and the restricted MATLAB execution environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 12:12 PM
Security Audit — agent-trust-hub — matlab-design-pcb-txline