Content
# 🦜🕸️ LangGraph Self-Paced Learning
[](https://langchain-ai.github.io/langgraph/)
[](https://www.python.org/)
[](https://deepmind.google/technologies/gemini/)
Welcome to the **Ultimate LangGraph 1.0 Course**! 🚀
This repository is designed to take you from a complete beginner to building production-ready **Autonomous Agents** using LangGraph and Google Gemini.
---
## 📚 How to Learn
This course follows a **"Learn by Doing"** philosophy. Each module is a self-contained mini-project.
1. **Concept (`concept.md`)**: Read this first to understand the *theory*.
2. **Project (`project.py`)**: Run this to see a *working example*.
3. **Challenge (`challenge.py`)**: A broken/incomplete script. **Your job is to fix it!**
4. **Solution (`solution.py`)**: Stuck? Check the reference answer.
---
## 🛠️ Setup
We use **[uv](https://github.com/astral-sh/uv)** for blazing fast dependency management.
1. **Install uv**:
```bash
# Windows
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
```
2. **Install Dependencies**:
```bash
uv sync
```
3. **Set API Keys**:
Rename `.env.example` to `.env` and add your **Google API Key**.
```bash
cp .env.example .env
# Edit .env and add GOOGLE_API_KEY=...
```
---
## 🗺️ Curriculum Overview
### 1. Foundations: Chain of Thought
**Folder:** `01-foundations`
Learn the basics of **Nodes**, **Edges**, and the **StateGraph**. We build a system that simulates a thinking process.
```mermaid
graph LR
START((Start)) --> Planner
Planner --> Analyst
Analyst --> Concluder
Concluder --> END((End))
style START fill:#f9f,stroke:#333,stroke-width:2px
style END fill:#f9f,stroke:#333,stroke-width:2px
```
* **Goal**: Pass a state dictionary through a sequence of functions.
* **Run**: `uv run 01-foundations/project.py`
---
### 2. State Management: Order Bot
**Folder:** `02-state-management`
Master the `State` schema. Learn how to use **Reducers** (like `add_messages`) and handle complex data flow.
```mermaid
graph LR
START((Start)) --> Validator
Validator --> Calculator
Calculator --> Receipt
Receipt --> END((End))
```
* **Goal**: Validate an order, calculate the total, and print a receipt.
* **Run**: `uv run 02-state-management/project.py`
---
### 3. Persistence: Support Bot
**Folder:** `03-persistence`
Add **Memory** to your agents! Learn about **Checkpointers** and **Thread IDs** to remember conversations across restarts.
```mermaid
graph TD
START((Start)) --> Bot
Bot --> END((End))
subgraph "💾 MemorySaver"
Bot
end
```
* **Goal**: A chatbot that remembers your name and issue even if you restart the script.
* **Run**: `uv run 03-persistence/project.py`
---
### 4. Human-in-the-Loop: Deployment Manager
**Folder:** `04-human-in-the-loop`
Learn how to **Pause** execution for human approval using `interrupt_before` and resume with `Command`.
```mermaid
graph TD
START((Start)) --> Prepare
Prepare -->|Interrupt| Deploy
Deploy --> END((End))
style Deploy stroke:#f00,stroke-width:4px
```
* **Goal**: A deployment system that waits for you to type "yes" before "deploying" code.
* **Run**: `uv run 04-human-in-the-loop/project.py`
---
### 5. Tool Use: Travel Agent
**Folder:** `05-tool-use`
Build a real **ReAct Agent**. Learn how to bind tools (functions) to an LLM and let it decide when to call them.
```mermaid
graph TD
START((Start)) --> Agent
Agent -->|Call Tool| Tools
Tools --> Agent
Agent -->|Final Answer| END((End))
```
* **Goal**: An agent that can search for flights (and weather in the challenge!).
* **Run**: `uv run 05-tool-use/project.py`
---
### 6. MCP Integration: Database Analyst
**Folder:** `06-mcp-integration`
Connect your agent to the world using the **Model Context Protocol (MCP)**. We simulate a connection to a SQLite database.
```mermaid
graph TD
START((Start)) --> Agent
Agent -->|MCP Protocol| SQLite[SQLite DB]
SQLite --> Agent
Agent --> END((End))
```
* **Goal**: An agent that can query a database to answer questions about users and orders.
* **Run**: `uv run 06-mcp-integration/project.py`
---
### 7. Special Project: A2A + MCP
**Folder:** `07-special-project-a2a-mcp`
Combine **Agent-to-Agent** communication with **MCP** tools.
- **Scout**: Finds the table.
- **Analyst**: Queries the table.
```mermaid
graph LR
START((Start)) --> Scout
Scout -->|Handoff| Analyst
Analyst --> END((End))
subgraph "Distributed Tools"
Scout -.-> ListTables
Analyst -.-> RunQuery
end
```
* **Goal**: A pipeline where one agent passes critical info to another to solve a task.
* **Run**: `uv run 07-special-project-a2a-mcp/project.py`
---
### 8. Multi-Agent Orchestration: Dev Team
**Folder:** `08-multi-agent-orchestration`
Scale up! Coordinate multiple specialized agents using the **Supervisor Pattern**.
```mermaid
graph TD
START((Start)) --> Supervisor
Supervisor -->|Route| Coder
Supervisor -->|Route| Tester
Coder --> Supervisor
Tester --> Supervisor
Supervisor -->|Finish| END((End))
```
* **Goal**: A "Supervisor" that manages a "Coder" and a "Tester" to write and verify code.
* **Run**: `uv run 07-multi-agent-orchestration/project.py`
---
### 9. Capstone: Autonomous Researcher
**Folder:** `09-capstone-project`
The Final Boss. Combine everything you've learned into a single, powerful application.
```mermaid
graph LR
START((Start)) --> Researcher
Researcher -->|Draft| Editor
Editor -->|Review| END((End))
subgraph "Tools"
Researcher -.-> Search
end
```
* **Goal**: An assistant that researches a topic, writes a draft, and asks for your approval before finalizing.
* **Run**: `uv run 08-capstone-project/main.py`
---
## 🤝 Contributing
Feel free to open issues or PRs if you find bugs or want to add more challenges!
Happy Coding! 🤖
