velt-suggestions-best-practices

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of documentation and code snippets providing implementation rules for Velt Suggestions in React and Next.js environments. No executable malicious code was found.
  • [CREDENTIALS_UNSAFE]: No hardcoded credentials or secrets were detected. The authentication examples correctly use the authProvider pattern with placeholders for API keys.
  • [EXTERNAL_DOWNLOADS]: The skill references the @veltdev/react package and official Velt documentation. These resources are standard vendor-provided components for the stated functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill documentation addresses workflows where AI agents can propose suggestions. It correctly identifies the attack surface of untrusted data by implementing a 'propose-then-review' lifecycle, ensuring that human reviewers must explicitly accept proposed changes before they are applied to the application state.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 09:30 PM
Security Audit — agent-trust-hub — velt-suggestions-best-practices