AI Agents, or Artificial Intelligence Agents, are systems that can act autonomously in a specific environment to achieve predetermined goals. These agents can learn and make decisions based on the feedback from the environment, optimizing their behavior and improving the efficiency of goal achievement. In different application scenarios, we design customized frameworks, using different large-scale models for different components in the application. These models can work together, achieving an effect where the whole is greater than the sum of its parts, i.e., “1+1>2”. This approach can help us create a real “hexagonal warrior”, which is a multi-faceted, high-performance AI system.