db-relational-design

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is restricted to design and documentation tasks. It does not use any tools, perform network requests, or execute commands on the host system.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of domain requirements. While this represents a standard injection surface, the skill has no access to sensitive capabilities or side-effect tools, rendering the risk associated with malicious inputs negligible.
  • [DATA_EXFILTRATION]: There are no patterns involving data retrieval or transmission to external domains. The skill explicitly states it is 'design only' and performs no database writes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 02:59 AM
Security Audit — agent-trust-hub — db-relational-design