mastra-ai-architecture-rules
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of instructional content and architectural rules formatted in Markdown. It does not contain any executable scripts, shell commands, or code.
- [SAFE]: No remote downloads, network exfiltration patterns, or obfuscation techniques were identified.
- [SAFE]: The instructions include proactive security guidelines, such as Rule 12 ('Security and Boundary Rules'), which explicitly warns against storing credentials in prompts or states and advises on the proper handling of sensitive file system access.
- [SAFE]: The skill promotes 'capability-aware' design, requiring explicit user approval before adding new infrastructure dependencies like databases, vector stores, or external packages.
Audit Metadata