AI Text Analysis Agent

A LangGraph-based AI agent that analyzes text documents by classifying them, extracting entities, and generating summaries.

📚 Attribution

This project is based on the tutorial "Your First AI Agent: Simpler Than You Think" by Nir Diamant.

Original Tutorial Resources:

• Tutorial Repository: agents-towards-production (8K+ ⭐)
• Author: Nir Diamant
• Tutorial Date: February 2025

📚 Attribution & Learning Journey

This project is based on the tutorial "Your First AI Agent: Simpler Than You Think" by Nir Diamant.

Learning Approach:

• Following tutorial with AI-assisted implementation (Claude)
• Focus: Understanding agent architecture, LangGraph workflows, and LLM integration
• Status: Learning project demonstrating AI agent fundamentals

Original Tutorial Resources:

• Tutorial Repository: agents-towards-production (8K+ ⭐)
• Author: Nir Diamant

Technical Updates Made:

• Updated imports for current LangChain version (langchain_core)
• Adapted setup for macOS development environment
• Added comprehensive documentation

🎯 Overview

This agent demonstrates core AI agent concepts including:

• State management - Maintains context throughout analysis
• Modular capabilities - Three specialized analysis nodes
• Graph-based workflow - Coordinated processing pipeline
• LLM integration - OpenAI GPT-4o-mini for intelligent processing

🏗️ Architecture

Input Text → Classification → Entity Extraction → Summarization → Results

The agent processes text through three sequential nodes:

1. Classification Node: Categorizes text (News, Blog, Research, Other)
2. Entity Extraction Node: Identifies people, organizations, locations
3. Summarization Node: Creates concise one-sentence summary

🚀 Quick Start

Prerequisites

• Python 3.8+
• OpenAI API key (Get one here)

Installation

1. Clone the repository

git clone https://github.com/YOUR_USERNAME/ai-text-analysis-agent.git
cd ai-text-analysis-agent

2. Create virtual environment

python3 -m venv agent_env
source agent_env/bin/activate  # On Windows: agent_env\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. Set up environment variables

cp .env.example .env
# Edit .env and add your OpenAI API key

5. Run the agent

python text_analysis_agent.py

📦 Dependencies

• langgraph - Graph-based agent workflow framework
• langchain - LLM orchestration
• langchain-openai - OpenAI integration
• python-dotenv - Environment variable management

💡 Usage Example

sample_text = """
OpenAI has announced the GPT-4 model, which exhibits 
human-level performance on various professional benchmarks...
"""

state_input = {"text": sample_text}
result = app.invoke(state_input)

print("Classification:", result["classification"])
print("Entities:", result["entities"])
print("Summary:", result["summary"])

Output:

Classification: News
Entities: ['OpenAI', 'GPT-4', 'GPT-3']
Summary: OpenAI's upcoming GPT-4 model is a multimodal AI 
that aims for human-level performance...

🛠️ Technical Implementation

State Management

Uses TypedDict to maintain structured state across nodes:

• text: Original input text
• classification: Document category
• entities: Extracted named entities
• summary: Generated summary

Node Functions

Each node is a pure function that:

• Receives current state
• Processes using LLM with specific prompt
• Returns updated state dict

Graph Workflow

LangGraph StateGraph coordinates execution:

• Sequential edge connections
• Automatic state merging
• Deterministic execution (temperature=0)

📊 Performance

• Model: GPT-4o-mini
• Average latency: ~3-5 seconds per analysis
• Cost: ~$0.003-$0.005 per analysis

🔮 Roadmap

• Add sentiment analysis node
• Implement conditional routing based on classification
• Add web search capability for fact-checking
• Support batch processing
• Add conversational memory
• Create REST API endpoint
• Deploy as web service

🤝 Contributing

Contributions welcome! Please feel free to submit a Pull Request.

📄 License

MIT License - See LICENSE file for details.

👤 Author

Adaugo Akaluso

• AI/Systems & Automation Designer
• GitHub: @adaakal
• LinkedIn: https://www.linkedin.com/in/adaugo-akaluso

🙏 Acknowledgments

• Built with LangGraph by LangChain
• Original tutorial by Nir Diamant
• Tutorial: "Your First AI Agent: Simpler Than You Think"
• OpenAI for GPT-4o-mini API

---

Learning AI Agent Development? Check out the original tutorial and Nir's repository with 25+ agent tutorials.