building-chatgpt-apps

Warn

Audited by Snyk on May 12, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.80). The skill instructs registering and using an external MCP server URL (example: https://abc123.ngrok-free.app/mcp) that ChatGPT will call at runtime and whose returned structuredContent and _meta/widget resources directly control the model's narration and widget behavior, so this is a runtime external dependency that can influence prompts.

Issues (1)

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
May 12, 2026, 11:56 AM
Issues
1