skills/qu-skills/superpowers/tools-ui/Gen Agent Trust Hub

tools-ui

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of UI components from https://ui.inference.sh/r/tools.json via the shadcn CLI. It also recommends the installation of external utility skills including belt-sh/cli and inference-sh/skills.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides UI components such as ToolCall and ToolResult which are designed to render data originating from AI tool executions. This creates an ingestion surface where potentially untrusted data from tool outputs is displayed in the user interface.
  • Ingestion points: result property in ToolResult and args property in ToolCall (found in SKILL.md).
  • Boundary markers: The components utilize React properties, which generally provide automatic escaping to prevent XSS.
  • Capability inventory: The skill context is limited to UI rendering; no code execution or file-system write capabilities are present in the analyzed scripts.
  • Sanitization: Sanitization is handled by the React framework's rendering engine for component properties.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:19 PM
Security Audit — agent-trust-hub — tools-ui