gate-frontend
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides conversational guidance for building frontend components. No malicious command execution, data exfiltration, or obfuscation was found.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied descriptions and project scaffolds to generate code, creating a potential attack surface.
- Ingestion points: User dialogue and project scaffold data.
- Boundary markers: None explicitly defined in the skill instructions.
- Capability inventory: Implicit file system write access for code generation.
- Sanitization: Relies on the underlying agent's default safety guardrails.
Audit Metadata