hello-ui

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues, malicious code, or suspicious patterns were detected in the skill instructions or logic.
  • [COMMAND_EXECUTION]: The skill utilizes a local utility script 'scripts/visual-state.mjs' to record visual validation status in a project-specific JSON file. This is a legitimate automation for tracking design consistency and does not involve remote code execution or privilege escalation.
  • [PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by ingesting project documentation. However, this is inherent to its functional requirements as a UI implementation tool. 1. Ingestion points: Project files 'plan.md', 'PRD', and '.helloagents/DESIGN.md' are used as primary sources of UI requirements. 2. Boundary markers: No specific boundary markers or instruction isolation techniques are used for the input documents. 3. Capability inventory: The skill can perform file system writes via local scripts and visual interactions using Playwright MCP. 4. Sanitization: Input documents are assumed to be trusted project artifacts and are not sanitized before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 01:43 PM
Security Audit — agent-trust-hub — hello-ui