Agentic Design Patterns(《智能体设计模式》),作者:Antonio Gulli
目录 – 共 424 页 = 1+2+1+1+4+9+103+61+34+114+74+5+4 11
Dedication(致谢词),1 页
Acknowledgment(致谢),2 页
Foreword(前言),1 页
A Thought Leader’s Perspective: Power and Responsibility(思想领袖的视角:权力与责任)
Introduction(导论),4 页
What makes an AI system an “agent”?(是什么让一个 AI 系统成为“智能体”?),9 页
Part One(第一部分,合计:103 页)
- Chapter 1: Prompt Chaining(提示链)(代码),12 页
- Chapter 2: Routing(路由)(代码),13 页
- Chapter 3: Parallelization(并行化)(代码),15 页
- Chapter 4: Reflection(反思/自检)(代码),13 页
- Chapter 5: Tool Use(工具使用)(代码),20 页
- Chapter 6: Planning(规划)(代码),13 页
- Chapter 7: Multi-Agent(多智能体)(代码),17 页
Part Two(第二部分,合计:61 页)
- Chapter 8: Memory Management(记忆管理)(代码),21 页
- Chapter 9: Learning and Adaptation(学习与自适应)(代码),12 页
- Chapter 10: Model Context Protocol(MCP,模型上下文协议)(代码),16 页
- Chapter 11: Goal Setting and Monitoring(目标设定与监控)(代码),12 页
Part Three(第三部分,合计:34 页)
- Chapter 12: Exception Handling and Recovery(异常处理与恢复)(代码),8 页
- Chapter 13: Human-in-the-Loop(人类在环)(代码),9 页
- Chapter 14: Knowledge Retrieval(RAG,检索增强生成)(代码),17 页
Part Four(第四部分,合计:114 页)
- Chapter 15: Inter-Agent Communication(A2A,智能体间通信)(代码),15 页
- Chapter 16: Resource-Aware Optimization(资源感知优化)(代码),15 页
- Chapter 17: Reasoning Techniques(推理技术)(代码),24 页
- Chapter 18: Guardrails / Safety Patterns(护栏/安全模式)(代码),19 页
- Chapter 19: Evaluation and Monitoring(评估与监测)(代码),18 页
- Chapter 20: Prioritization(优先级管理)(代码),10 页
- Chapter 21: Exploration and Discovery(探索与发现)(代码),13 页
Appendix(附录,合计:74 页)
- Appendix A: Advanced Prompting Techniques(高级提示工程技术),28 页
- Appendix B: AI Agentic ……:From GUI to Real-World Environment(从图形界面到真实世界环境),6 页
- Appendix C: Quick Overview of Agentic Frameworks(智能体框架速览),8 页
- Appendix D: Building an Agent with AgentSpace(在线),6 页
- Appendix E: AI Agents on the CLI(命令行中的 AI 智能体,在线),5 页
- Appendix F: Under the Hood: An Inside Look at the Agents’ Reasoning Engines(幕后:走近智能体推理引擎内部),14 页
- Appendix G: Coding Agents(编写智能体),7 页
Conclusion(结论),5 页
Glossary(术语表),3 页