Security and ethical guardrails for an AI workflow

You are working on the integration of an LLM into a development workflow that helps automate coding tasks.

Before deploying the system, you need to review sample prompts that users might submit to the AI assistant. This review is part of designing the security and ethical guardrails for the workflow, ensuring the AI doesn't produce or assist in generating harmful or unethical code.

Which of the following prompts could raise ethical concerns, security risks, or violate acceptable use policies?