ai-visual-prompt-review

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

• Potentially malicious external URL detected (high risk: 0.90). The runner (scripts/run.py) makes runtime POST requests to the remote execution endpoint https://ai-skills.ai/api/execute (DEFAULT_BASE_URL + /api/execute), which executes the skill remotely and therefore directly controls agent execution and prompts.