✨
AI Summary
- AI agents executing commands on developer machines pose security risks.
- Prompt injection via markdown files can lead to malicious code execution.
- Sarah Sanders emphasizes the need for secure agent execution environments.