🤖 LTM-AI-Agent

LTM-AI-Agent is an AI chatbot with Long-Term Memory (LTM) that remembers user interactions, recalls past conversations, and adapts responses based on stored context.

🚀 Features

• ✅ Persistent Memory: Stores & retrieves chat history from SQLite.
• ✅ Context Awareness: Remembers user details across sessions.
• ✅ Contradiction Handling: Updates facts when new information is provided.
• ✅ Inference & Reasoning: Understands and adapts responses based on memory.
• ✅ Time-Based Memory: Can recall past events over time.

🛠 Tech Stack

• Python 🐍
• SQLite 🗄️
• Pydantic AI 🤖
• OpenAI GPT-3.5 Turbo

📂 Project Structure

📦 LTM-AI-Agent
├── 📜 store_conversation_advance.py  # Chatbot logic & memory handling
├── 📜 conversation1.db               # SQLite database (auto-created)
├── 📜 requirements.txt               # Dependencies
├── 📜 README.md                      # Project documentation

🚀 Clone & Run

1️⃣ Clone the repository

git clone https://github.com/gomesjonathan99/LTM-AI-Agent.git

2️⃣ Install dependencies

pip install -r requirements.txt

3️⃣ Run the chatbot

python store_conversation_advance.py

📌 Usage

• Start a conversation

  print(chat(session_id="session_12", message="Hello!"))

• Retrieve past knowledge

  print(chat(session_id="session_12", message="What do you remember about me?"))

• Handle contradictions

  print(chat(session_id="session_12", message="Actually, I don't have a pet."))
  print(chat(session_id="session_12", message="Do I have a pet?"))

---