add-component

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a collection of UI templates and integration instructions. No malicious patterns, obfuscation, or unauthorized access attempts were detected.
  • [PROMPT_INJECTION]: The skill provides components for handling user input, which represents a surface for indirect prompt injection.
  • Ingestion points: Multiple templates (e.g., GlobalSearch.tsx, filter-panel.tsx, and various form components) ingest untrusted data from user queries and inputs.
  • Boundary markers: The skill's documentation recommends using boundary markers and explicit instructions to ignore untrusted content when processing data.
  • Capability inventory: The provided components are purely presentational and do not execute sensitive operations directly. They are intended to be wired to secure backend functions.
  • Sanitization: The skill emphasizes best practices like Zod schema validation and server-side authorization checks (requireUser, requireRole) to mitigate risks from untrusted data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 06:44 AM
Security Audit — agent-trust-hub — add-component