▣ Chapter 1: Agents and Their Ecosystems
1.1 Definition of Agent
1.2 Agent Components
1.3 Why Agents Are Getting Attention
1.4 The World Inside the AI ​​Interface
1.5 Beginning your journey into the world of AI agents
summation

▣ Chapter 2: Demonstrating the Power of LLM
2.1 Working with the OpenAI API
__2.1.1 Connecting to the Conversation Completion Model
__2.1.2 Understanding Requests and Responses
2.2 Exploring and Leveraging Open Source LLMs Using LM Studio
__2.2.1 Installing and Running LM Studio
__2.2.2 Serving LLM locally with LM Studio
2.3 LLM Prompting through Prompt Engineering
__2.3.1 Detailed query
__2.3.2 Persona Adoption
__2.3.3 Using delimiters
__2.3.4 Step Specification
__2.3.5 Example Tactics
__2.3.6 Specifying output length
2.4 Choosing the LLM that best suits your specific needs
2.5 Practice Problems
summation

▣ Chapter 3: Using the GPT Assistant
3.1 Exploring OpenAI Assistants with ChatGPT
3.2 Creating a GPT that acts as a data scientist
3.3 Customizing GPT and Adding Custom Actions
__3.3.1 Creating an Assistant to Help Build Assistants
__3.3.2 Connecting Custom Actions to Assistants
3.4 Expanding Assistant Knowledge Using File Uploads
__3.4.1 Creating an 'Easy Calculus' GPT
__3.4.2 Added knowledge retrieval and reference capabilities using file uploads
3.5 GPT Post
__3.5.1 Expensive GPT Assistant
__3.5.2 Understanding the Economics of GPT
__3.5.3 GPT Publishing and Sharing
3.6 Practice Problems
summation

▣ Chapter 4: Examining Multi-Agent Systems
4.1 Introduction to Multi-Agent Systems through AutoGen Studio
__4.1.1 AutoGen Studio Installation and Usage
__4.1.2 Adding Skills in Autogen Studio
4.2 Using the AutoGen Library
__4.2.1 Autogen Installation and Utilization
__4.2.2 Added a critique agent to improve code output
__4.2.3 Understanding Autogen Cache
4.3 Group Chat Using Agents and Autogen
4.4 Building an Agent Crew Using Crew AI
__4.4.1 Creating a 'Wacky' Agent Crew with Crew AI
__4.4.2 Observing agent operation using AgentOps
4.5 Coding Agents Revisited with CrewAI
4.6 Practice Problems
summation

▣ Chapter 5: Strengthening Agent Functions through Actions
5.1 Definition of Agent Behavior
5.2 Defining and Executing OpenAI Functions
__5.2.1 Add function to LLM API call
__5.2.2 Execution of function call behavior
5.3 Introducing the Semantic Kernel
__5.3.1 Getting Started with SK Semantic Functions
__5.3.2 Semantic Functions and Context Variables
5.4 Synergy between semantic functions and native functions
__5.4.1 Creating and Registering Semantic Skills/Plugins
__5.4.2 Applying native functions
__5.4.3 Embedding native functions into semantic functions
5.5 Semantic Kernel as an Interactive Service Agent
__5.5.1 Building a Semantic GPT Interface
__5.5.2 Semantic Service Testing
__5.5.3 Interactive Chat Using the Semantic Service Layer
5.6 Creating a Semantic Service Considering LLM's Semantic Understanding Ability
5.7 Practice Problems
summation

▣ Chapter 6: Building an Autonomous Assistant
6.1 Introduction to the Behavior Tree
__6.1.1 Executing the Behavior Tree
__6.1.2 Advantages of Behavior Trees
__6.1.3 Implementing a Behavior Tree Using Python and py_trees
6.2 Explore the GPT Assistants Playground
__6.2.1 Installing and Running the Playground
__6.2.2 Using and Creating Custom Behaviors
__6.2.3 Installing the Assistants Database
__6.2.4 To make the assistant run code locally
__6.2.5 Investigating Assistant Processes Using Logs
6.3 Introduction to Agent-Based Behavior Trees (ABTs)
__6.3.1 Managing Assistants as Assistants
__6.3.2 Creating a Coding Challenge ABT
__6.3.3 Comparison of Conversational AI Systems with Other Methods
__6.3.4 Posting YouTube videos to Twitter (now X)
__6.3.5 Required Twitter (now X) settings
6.4 Building a Conversational Autonomous Multi-Agent System
6.5 Building an ABT using a back chain
6.6 Practice Problems
summation

▣ Chapter 7: Building and Utilizing an Agent Platform
7.1 Introducing Nexus: Not Just Another Platform
Running __7.1.1 Nexus
__7.1.2 Nexus Development Mode
7.2 Introducing Streamlet for Chat Application Development
__7.2.1 Creating a Streamlet Chat Application
__7.2.2 Creating a Streaming Chat Application
7.3 Developing Agent Profiles and Personas
7.4 Agent Engine that Drives Agents
7.5 Giving Agents Actions and Tools
7.6 Practice Problems
summation

▣ Chapter 8: Agent's Memory and Knowledge
8.1 The Meaning and Importance of Search in AI Applications
8.2 Basic Principles of RAG
8.3 Details of semantic search and document indexing
__8.3.1 Application of vector similarity search
__8.3.2 Similarity Search with Vector Databases
__8.3.3 Understanding Document Embedding
__8.3.4 Document Embedding Search Using Chroma DB
8.4 Building RAG using Langchain
__8.4.1 Document Splitting and Loading Using Lang Chain
__8.4.2 Token-unit document splitting using Langchain
8.5 Applying RAG to Building Agent Knowledge
8.6 Memory Implementation in Agent-Type Systems
__8.6.1 Using Nexus's memory storage
__8.6.2 Semantic memory and its applications
8.7 Compression of Memory and Knowledge
8.8 Practice Problems
summation

▣ Chapter 9: Effective Agent Prompting Using Prompt Flow
9.1 Why Systematic Prompt Engineering is Needed
9.2 Understanding Agent Profiles and Personas
9.3 Setting up the initial prompt flow
__9.3.1 Getting Started
__9.3.2 Creating a profile with a Jinja2 template
__9.3.3 Deploying the Prompt Flow API
9.4 Profile Evaluation: Rubrics and Grounding
9.5 Rubrics and Grounding
9.6 Grounding Assessment Using the LLM Profile
9.7 Comparing Multiple Profiles: Getting the Perfect Profile
__9.7.1 Parsing LLM Evaluation Output
__9.7.2 Batch execution of prompt flow
__9.7.3 Creating a Grounding Evaluation Flow
9.8 Practice Problems
summation

▣ Chapter 10: Agent Reasoning and Evaluation
10.1 Understanding Direct Solution Prompting
__10.1.1 Question and Answer Prompting
__10.1.2 Few-shot prompting
__10.1.3 Generality Extraction Using Zero-Shot Prompting
10.2 Prompt Engineering and Inference
__10.2.1 Incident Chain Prompting
__10.2.2 Zero-shot CoT prompting
__10.2.3 Step-by-step prompt chaining
10.3 Using Assessments for Consistent Answers
__10.3.1 Evaluation of Self-Consistency Prompting
__10.3.2 Evaluation of Thought Tree Prompting
10.4 Practice Problems
summation

▣ Chapter 11: Agent Planning and Feedback
11.1 Planning: Essential Tools for Every Agent/Assistant
11.2 Sequential Planning Process
11.3 Building a Sequential Planner
11.4 Step-by-Step Planner Review: OpenAI's Inference-Specific Model
11.5 The Uses and Applications of Planning, Inference, Evaluation, and Feedback in Assistant and Agent-Type Systems
__11.5.1 Purpose and Use of the Plan
__11.5.2 Uses and Usage of Inference
__11.5.3 Purpose and usage of evaluation
__11.5.4 Uses and Usage of Feedback
11.6 Practice Problems
summation

▣ Appendix A: Utilizing OpenAI LLM
A.1 Creating an OpenAI Account and Key
A.2 Azure OpenAI Studio API Key and Distribution

▣ Appendix B: Python Development Environment
B.1 Download the example code
B.2 Installing Python
B.3 Installing and Setting Up VS Code
B.4 Installing VS Code Extensions for Python Development
B.5 Creating a New Python Environment with VS Code
B.6 Using Containers (Docker) with the Dev Containers Extension

★ What this book covers ★

◎ Understand and implement AI agent behavior patterns
◎ Design and deploy intelligent agents that can actually be operated.
◎ Using the OpenAI Assistants API and complementary tools
◎ Implementing a robust knowledge management and memory system
◎ Creating self-improving agents through feedback loops
◎ Configuring a collaborative multi-agent system
◎ Enhance your agents with voice and vision capabilities