ux-review

Warn

Audited by Snyk on Feb 20, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs agents to open and crawl the user's provided target URL and follow navigation links using Playwright CLI (see "Phase 2: Discovery" and the agent workflows in SKILL.md which call playwright-cli open <url> and snapshot/follow pages), meaning it fetches and reads arbitrary third‑party web content that directly drives agent actions and decisions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 20, 2026, 06:06 AM