iso-13485-certification
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely within the user's local environment. The provided Python script (
scripts/gap_analyzer.py) uses standard libraries to perform keyword searches in local documents and generates a static JSON report. It contains no network operations, dynamic code execution, or data exfiltration logic. - [SAFE]: Documentation templates and guides are standard markdown files intended for manual or agent-assisted customization using simple string replacement for placeholders (e.g., '[COMPANY NAME]'). No obfuscation or hidden instructions were found.
- [SAFE]: Indirect Prompt Injection analysis: While the skill processes user-provided documentation, it does so using a static script that performs keyword matching rather than direct LLM interpretation of untrusted data in an executable context. No evidence of autonomous execution or bypass of safety filters was detected.
Audit Metadata