This research introduces a comprehensive security architecture for LLM-driven robotic agents conducting online financial transactions, addressing critical vulnerabilities unique to these systems.

• Implements multi-layered security including blockchain verification, multi-factor authentication, and real-time anomaly detection
• Demonstrates significant reduction in transaction fraud and improved breach detection metrics
• Establishes security protocols specifically designed for autonomous AI agents operating in e-commerce environments

This work addresses the urgent security challenges as LLM-powered robots increasingly handle sensitive financial transactions in commercial settings, providing a practical framework for organizations deploying these technologies.

Enforcing Cybersecurity Constraints for LLM-driven Robot Agents for Online Transactions