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
Risk Level
SAFE
Analyzed
Sep 19, 2026, 04:26 AM
Security Audit — agent-trust-hub — gate-frontend