grdb

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • PROMPT_INJECTION (SAFE): No override markers, role-play injections, or system prompt extraction attempts were found. The instructions are purely technical and aimed at guiding the agent's code generation for database operations.
  • DATA_EXPOSURE & EXFILTRATION (SAFE): The skill uses standard iOS/macOS file system paths for database storage (Documents directory). No hardcoded credentials, sensitive system paths (like SSH keys), or suspicious network operations were detected.
  • REMOTE_CODE_EXECUTION (SAFE): There are no patterns of piping remote scripts to shells (curl|bash). The skill references the legitimate open-source repository https://github.com/groue/GRDB.swift for installation via Swift Package Manager.
  • INDIRECT PROMPT INJECTION (SAFE): While the skill enables the agent to execute SQL queries, the provided documentation explicitly encourages the use of parameterized queries and type-safe interfaces, which are industry-standard defenses against SQL injection.
  • DYNAMIC EXECUTION (SAFE): The skill focuses on static code patterns and standard library usage. No unsafe deserialization (pickle) or runtime code generation from untrusted sources was found.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:49 PM