skills/mblode/agent-skills/ui-design/Gen Agent Trust Hub

ui-design

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a high-quality UI/UX assistant. Analysis across all 11 threat categories confirms no malicious behavior.
  • [PROMPT_INJECTION]: Instructions focus on design and development constraints (e.g., 'Never use emojis in headings'). There are no attempts to override agent safety filters or system prompts.
  • [DATA_EXFILTRATION]: Network operations are limited to fetching placeholder assets from the vendor-owned domain assets.ui.sh and reputable font providers (rsms.me, macpaw.com, fontshare.com). No credential harvesting or sensitive file access was identified.
  • [REMOTE_CODE_EXECUTION]: Instructions involve standard developer workflows like using the Tailwind CSS CLI (pnpm dlx @tailwindcss/cli) for class canonicalization. No execution of untrusted remote scripts was found.
  • [COMMAND_EXECUTION]: The Audit mode uses shell commands like rg (ripgrep) and git diff to identify patterns in the user's codebase. These are used strictly for analysis purposes as defined in the rule files.
  • [INDIRECT_PROMPT_INJECTION]: The skill allows users to suppress specific rules using comments like // ui-audit-ignore:<rule-id>. This is a documented feature for handling false positives in code audits and does not constitute a vulnerability.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 10:59 PM