skills/affaan-m/ecc/ui-demo/Gen Agent Trust Hub

ui-demo

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions and a JavaScript template for creating a Node.js script (demo-script.cjs) and executing it via the node command to perform browser automation tasks.
  • [COMMAND_EXECUTION]: It uses page.evaluate() to dynamically inject and execute JavaScript code within the browser context to facilitate element discovery and UI overlays.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and displays text content from interactive elements on arbitrary web pages in the Discovery phase.
  • Ingestion points: page.evaluate in SKILL.md.
  • Boundary markers: Absent.
  • Capability inventory: Full browser control via Playwright and file system writes via fs.
  • Sanitization: None identified for the ingested text.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 02:48 PM
Security Audit — agent-trust-hub — ui-demo