fix small bug in chapter 6

docs/chapter6/Chapter6-Framework-Development-Practice.md

@@ -28,7 +28,7 @@ Therefore, moving from manual implementation to framework development is not onl
 
 The ecosystem of agent frameworks is developing at an unprecedented speed. If LangChain and LlamaIndex defined the paradigm of the first generation of general LLM application frameworks, then the new generation of frameworks is more focused on solving deep challenges in specific domains, especially **Multi-Agent Collaboration** and **Complex Workflow Control**.
 
-In the subsequent practical work of this chapter, we will focus on four frameworks that are highly representative in these cutting-edge fields: AutoGen, AgentScope, CAMEL, and LangGraph. Their design philosophies are different, representing different technical paths for implementing complex agent systems, as shown in Figure 6.1.
+In the subsequent practical work of this chapter, we will focus on four frameworks that are highly representative in these cutting-edge fields: AutoGen, AgentScope, CAMEL, and LangGraph. Their design philosophies are different, representing different technical paths for implementing complex agent systems, as shown in Table 6.1.
 
 <div align="center">
   <p>Table 6.1 Comparison of Four Agent Frameworks</p>

docs/chapter6/第六章 框架开发实践.md

@@ -28,7 +28,7 @@
 
 智能体框架的生态正在以前所未有的速度发展。如果说 LangChain 和 LlamaIndex 定义了第一代通用 LLM 应用框架的范式，那么新一代的框架则更加专注于解决特定领域的深层挑战，尤其是<strong>多智能体协作 (Multi-Agent Collaboration)</strong> 和 <strong>复杂工作流控制 (Complex Workflow Control)</strong>。
 
-在本章的后续实战中，我们将聚焦于四个在这些前沿领域极具代表性的框架：AutoGen、AgentScope、CAMEL 和 LangGraph。它们的设计理念各不相同，分别代表了实现复杂智能体系统的不同技术路径，如图6.1所示。
+在本章的后续实战中，我们将聚焦于四个在这些前沿领域极具代表性的框架：AutoGen、AgentScope、CAMEL 和 LangGraph。它们的设计理念各不相同，分别代表了实现复杂智能体系统的不同技术路径，如表6.1所示。
 
 <div align="center">
   <p>表 6.1 四种智能体框架对比</p>