firestore-rules-creation
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill enforces robust security practices, including mandatory helper functions for authentication and ownership, and a structured workflow that prioritizes data integrity and PII protection. It requires an adversarial testing phase ('Devil's Advocate') to confirm rule security.
- [INDIRECT_PROMPT_INJECTION]: The skill scans codebase files (Ingestion point) to identify schemas. While boundary markers are absent, the capability inventory includes CRUD operation identification and rule generation. The mandatory Phase-3 'Devil's Advocate Attack' serves as a sanitization step to ensure the final output is secure despite untrusted inputs.
- [COMMAND_EXECUTION]: The skill mentions using standard vendor tools for syntax validation, which is a benign and expected use case for this type of deployment assistant.
Audit Metadata