tools-ui

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of UI components from ui.inference.sh using the shadcn CLI tool. This is a standard workflow for integrating remote component registries in modern web development.
  • [INDIRECT_PROMPT_INJECTION]: The UI components serve as an interface for tool metadata and results, creating an ingestion point for external data. 1. Ingestion points: Data enters through the name, args, and result props of components like ToolCall, ToolResult, and ToolApproval (SKILL.md). 2. Boundary markers: No explicit isolation markers are defined for displayed content. 3. Capability inventory: The skill consists solely of UI rendering logic and does not include capabilities for file system access, network exfiltration, or command execution. 4. Sanitization: Relies on standard React rendering to prevent XSS. The surface is not exploitable due to the restricted capability scope.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 11:00 PM
Security Audit — agent-trust-hub — tools-ui