🧠 Model-Based Agent

A Python implementation of a Model-Based AI Agent with memory and state management.

---

🤖 What is a Model-Based Agent?

A Model-Based Agent can:

✅ remember past information ✅ use memory to make decisions ✅ maintain internal state

Unlike a Simple Reflex Agent, this agent does not rely only on current input.

---

🏗️ Architecture

        +------------------+
        |   User Input     |
        +--------+---------+
                 |
                 v
        +------------------+
        |  Input Cleaner   |
        +--------+---------+
                 |
                 v
        +------------------+
        | Information      |
        | Extraction       |
        +--------+---------+
                 |
                 v
        +------------------+
        |     Memory       |
        +--------+---------+
                 |
                 v
        +------------------+
        | Decision Engine  |
        +--------+---------+
                 |
                 v
        +------------------+
        |     Output       |
        +------------------+

---

🔄 Agent Flow

User → main.py → agent.py → processor.py → memory.py → Response

---

📂 Project Structure

model-based-agent/
│
├── agent/
│   ├── memory.py
│   ├── processor.py
│   ├── rules.py
│   └── agent.py
│
├── main.py
├── README.md
└── .gitignore

---

⚙️ Features

✅ Stores user information ✅ Retrieves stored memory ✅ Handles multiple memory types ✅ Modular architecture ✅ Stateful behavior

---

🧠 Memory Capabilities

The agent can remember:

• user name
• location
• future expandable memory fields

---

🚀 How to Run

1️⃣ Create Virtual Environment

python -m venv venv

2️⃣ Activate Environment

Windows

venv\Scripts\activate

Mac/Linux

source venv/bin/activate

3️⃣ Run Agent

python main.py

---

🧪 Example Interaction

You: my name is Ravi
Agent: Got it! I will remember your name is Ravi

You: what is my name
Agent: Your name is Ravi

You: i live in Chennai
Agent: Got it! I will remember your location is Chennai

You: where do i live
Agent: You live in Chennai

---

🧠 Core Concepts Learned

• Stateful agents
• Memory systems
• Information extraction
• Internal state management
• Modular AI architecture

---

⚠️ Limitations

• No learning
• No planning
• No reasoning
• Only remembers predefined information

---

🚀 Future Improvements

• Persistent memory
• Database integration
• Planning systems
• Multi-step reasoning
• Learning mechanisms

---

📚 Learning Outcome

This project demonstrates:

Input + Memory → Decision → Action

---

🔥 Key Upgrade from Simple Reflex Agent

Feature	Simple Reflex	Model-Based
Memory	❌	✅
State	❌	✅
Context Awareness	❌	✅
Rule-Based	✅	✅

---