🤖 Learn Agent Development from Scratch

A systematic, comprehensive, and practice-oriented AI Agent development guide

Daily auto-tracking of arXiv frontier papers — content stays cutting-edge, always.

   

🐛 Report Issues · 💬 Discussions · 🇨🇳 中文版 README

---

🚀 Auto-Tracking Frontier: Daily arXiv Paper Updates

🤖 This repository automatically searches arXiv for the latest AI Agent-related papers every day and updates the content accordingly — ensuring you always stay at the cutting edge of research!

• 📡 Daily Automated Search: A scheduled pipeline scans arXiv daily for new papers on Agent architectures, tool use, memory systems, multi-agent collaboration, reinforcement learning for agents, and more.
• 📝 Auto-Updated Content: Relevant findings are automatically integrated into the corresponding chapters, keeping the book's frontier sections fresh and up-to-date.
• 🔔 Never Miss a Breakthrough: No need to manually track dozens of research feeds — this repo does it for you, so you can focus on learning and building.

💡 This means the content you read here is not static — it evolves continuously with the latest advances in the AI Agent field.

---

✨ Key Features

• 🎯 Step by Step: From LLM fundamentals to multi-Agent systems, each chapter has a clear knowledge progression
• 💻 Code First: Every core concept comes with runnable Python code examples
• 🎨 Rich Illustrations: 120+ hand-drawn SVG architecture diagrams / flowcharts / sequence diagrams for intuitive understanding
• 🎬 Interactive Animations: 5 built-in interactive HTML animations (Perceive-Think-Act cycle, ReAct reasoning, Function Calling, RAG flow, GRPO sampling)
• 🔬 Paper Reviews: Key chapters include frontier paper deep-dives (ReAct, Reflexion, MemGPT, GRPO, etc.)
• 🏗️ Complete Projects: 3 comprehensive hands-on projects (AI Coding Assistant, Intelligent Data Analysis Agent, Multimodal Agent)
• 🛡️ Production Ready: Covers security, evaluation, deployment, and other production essentials
• 🧪 Cutting Edge: Covers Context Engineering, Agentic-RL (GRPO/DPO/PPO), MCP/A2A/ANP, and other 2025–2026 latest advances
• 📐 Formula Support: KaTeX-rendered math formulas for clear reading of policy gradient, KL divergence derivations in RL chapters
• 🔄 Continuously Updated: Tracking the latest changes in LangChain, LangGraph, MCP, and other frameworks

---

📸 Selected Content Preview

Below are selected showcases from the book's 120+ hand-drawn SVG illustrations, all original to this book.

🧠 Agent Core Architecture

Perceive-Think-Act Loop (Chapter 1) Agent's core mechanism: Perceive environment → LLM reasoning → Execute action → Loop until goal achieved

ReAct Reasoning Framework (Chapter 6) Thought → Action → Observation alternating loop, enabling Agents to think while acting

🛠️ Tool Calling & RAG

Function Calling Complete Flow (Chapter 4) 6-step complete flow from user input to tool invocation to final response, with message structure illustration

RAG Retrieval-Augmented Generation (Chapter 7) Offline indexing + Online retrieval dual-phase architecture, making LLM answers evidence-based

💾 Memory System & Context Engineering

Three-Layer Memory Architecture (Chapter 5) Working memory → Short-term memory → Long-term memory, with important info sinking down and semantic retrieval pulling up

Prompt Engineering vs Context Engineering (Chapter 8) From "how to say it" to "what the LLM sees" — the paradigm shift of the Agent era

🤝 Multi-Agent & Communication Protocols

Three Multi-Agent Communication Patterns (Chapter 14) Message Queue (async decoupling) / Shared Blackboard (data sharing) / Direct Call (real-time collaboration)

MCP / A2A / ANP Protocol Comparison (Chapter 15) Three-layer protocol stack: ANP for discovery → A2A for task collaboration → MCP for tool invocation

🧪 Reinforcement Learning & Frameworks

GRPO Training Architecture (Chapter 10) No Critic model needed, computes advantage via intra-group normalization, only 1.5× model size in VRAM

LangGraph Three Core Concepts (Chapter 12) State (shared state) · Node (processing unit) · Edge (execution flow control)

📖 The above is just a selected preview — For the full 120+ architecture diagrams + 5 interactive animations, please read online

---

🎬 Interactive Animations

This book includes 5 interactive HTML animations to help you intuitively understand the dynamic processes of core concepts:

Animation	Chapter	Description
🔄 Perceive-Think-Act Cycle	Chapter 1	Dynamic demonstration of Agent's core loop
💡 ReAct Reasoning Process	Chapter 6	Shows the alternating Thought → Action → Observation process
🔧 Function Calling	Chapter 4	Complete tool invocation flow animation
📚 RAG Retrieval Flow	Chapter 7	From document chunking to vector retrieval to answer generation
🎯 GRPO Sampling Process	Chapter 10	Visualization of intra-group multi-output sampling and reward normalization

💡 Interactive animations are only available in the online e-book. Local builds can also preview them.

---

🔥 Core Topics at a Glance

🧠 Agent Core Architecture Perceive → Think → Act Loop ReAct Reasoning Framework Task Decomposition & Planning Reflection & Self-Correction 🛠️ Tools & Skills Function Calling Mechanism Custom Tool Design Skill System Construction Tool Description Best Practices 🧪 Reinforcement Learning Training SFT + LoRA Basic Training PPO / DPO / GRPO Algorithm Deep-Dive Complete Training Pipeline Hands-on 2025–2026 Latest Research Advances

💾 Memory, Knowledge & Context Short-term / Long-term / Working Memory Vector Databases (Chroma / FAISS) RAG Retrieval-Augmented Generation Context Engineering & Attention Budget 🤝 Multi-Agent Collaboration & Communication MCP / A2A / ANP Protocol Stack Supervisor vs Decentralized Patterns CrewAI / AutoGen Frameworks LangGraph Stateful Agents 🛡️ Production Full Pipeline Evaluation Benchmarks (GAIA / SWE-bench) Security Defense & Sandbox Isolation Containerized Deployment & Streaming Observability & Cost Optimization

---

🚀 Quick Start

Local Build

# Install mdBook (choose one)
cargo install mdbook
# Or macOS: brew install mdbook

# Install mdbook-katex plugin (for math formula rendering)
cargo install mdbook-katex

# Clone the repository
git clone https://github.com/Haozhe-Xing/agent_learning.git
cd agent_learning

# Build both Chinese and English versions and start unified server (default port 3000)
./serve.sh

After starting, visit:

• 🌐 Language Selection Home: http://localhost:3000
• 🇨🇳 Chinese Version: http://localhost:3000/zh/
• 🇺🇸 English Version: http://localhost:3000/en/

Environment Setup (For Code Practice)

# Python 3.11+
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install core dependencies
pip install langchain langchain-openai langgraph openai anthropic

# Configure API Key
export OPENAI_API_KEY="your-key-here"

---

📊 Technology Stack

---

🤝 Contributing

All forms of contribution are welcome!

• 🐛 Found a bug: Submit an Issue
• 💡 Content suggestions: Start a Discussion
• 📝 Improve content: Fork → Edit → Submit PR
• ⭐ Support the project: Give this repo a Star!

Contributing Guide

# Fork and clone
git clone https://github.com/YOUR_USERNAME/agent_learning.git

# Create a feature branch
git checkout -b feature/improve-chapter-4

# Local preview
./serve.sh

# Commit and push
git commit -m "feat: improve Chapter 4 tool calling code examples"
git push origin feature/improve-chapter-4

Content Organization Conventions

• Each chapter is placed in a separate directory src/zh/chapter_xxx/ (Chinese) or src/en/chapter_xxx/ (English)
• Chapter overview goes in README.md, sections are numbered as 01_xxx.md, 02_xxx.md
• Chinese SVG illustrations go in src/zh/svg/, English versions in src/en/svg/, naming format: chapter_xxx_description.svg
• Chinese interactive animations go in src/zh/animations/, English versions in src/en/animations/

---

📄 License

This project is open-sourced under the MIT License.

---

⭐ Star History

If this project helps you, please give it a Star ⭐ — it's the greatest encouragement for the author!

---

Built with ❤️, so that every developer can master AI Agent development

⬆ Back to Top