gsp-backend-selector
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions were found to be free of malicious patterns, such as prompt injection, obfuscation, or data exfiltration. The skill is designed for architectural decision-making within a local development context.
- [INDIRECT_PROMPT_INJECTION]: The skill surface includes ingestion of external local data which could potentially influence agent behavior.
- Ingestion points: Reads backend capabilities from
../../shared/reference/backend-profiles.md. - Boundary markers: None are defined to separate the reference data from instructions.
- Capability inventory: The skill writes analysis to
docs/game-studio/backend-decision.mdbut lacks network access or arbitrary command execution tools. - Sanitization: No validation or sanitization of the reference file content is mentioned.
Audit Metadata