frontend-workflow
Warn
Audited by Snyk on Mar 11, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 1.00). The skill mandates runtime calls to the external Frontend MCP server (e.g., MCP tool endpoints such as fetch_tailwind_docs(...), fetch_alpine_patterns(...), fetch_vite_config(...), fetch_typescript_config(...), get_ecommerce_ui_pattern(...)) to retrieve documentation which the agent must follow, so externally-fetched content directly controls the agent's prompts/instructions at runtime.
Audit Metadata