README.md

---
title: Competitive Analysis Single Agent
emoji: 📊
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 4.19.2
python_version: 3.11
app_file: app.py
pinned: false
---

# 🏆 Competitive Analysis Agent - MCP Implementation

A sophisticated single-agent system for comprehensive competitive analysis using the **Model Context Protocol (MCP)** with FastMCP. This implementation combines web research, intelligent reasoning, and structured analysis to provide actionable competitive insights.

## 🎯 Architecture Overview

```
┌──────────────────────────────────────────────────────────┐
│     Single-File Application (app.py)                     │
├──────────────────────────────────────────────────────────┤
│                                                          │
│  ┌────────────────────────────────────────────────────┐ │
│  │  Gradio Web Interface (User UI)                    │ │
│  │            :7860                                   │ │
│  └────────────────────────────────────────────────────┘ │
│                        │                                │
│  ┌────────────────────────────────────────────────────┐ │
│  │  Analysis Engine & Tools:                          │ │
│  │  • validate_company()       - Verify company      │ │
│  │  • identify_sector()        - Find industry       │ │
│  │  • identify_competitors()   - Discover rivals     │ │
│  │  • browse_page()            - Extract data        │ │
│  │  • generate_report()        - Create analysis     │ │
│  └────────────────────────────────────────────────────┘ │
│                        │                                │
│  ┌────────────────────────────────────────────────────┐ │
│  │  AI Agent (OpenAI GPT-4)                           │ │
│  │  Strategic reasoning & insights generation        │ │
│  └────────────────────────────────────────────────────┘ │
│                                                          │
│              External APIs:                             │
│              • DuckDuckGo (Web Search)                 │
│              • OpenAI (LLM Analysis)                   │
│              • External Web Pages (Scraping)           │
│                                                          │
└──────────────────────────────────────────────────────────┘
```

**Single-File Design**: All functionality is consolidated in one `app.py` file for simplicity and ease of deployment. No separate MCP server needed.

## ✨ Features

### 🔍 Company Analysis
- **Validation**: Confirms company legitimacy using web search
- **Sector Identification**: Determines industry using multi-stage analysis
- **Competitor Discovery**: Identifies top 3 competitors with ranking

### 📊 Strategic Analysis
- **Web Research**: Automatically gathers company information
- **Content Extraction**: Extracts relevant strategic data
- **Structured Reports**: Generates professional analysis with:
  - Executive summary
  - Competitor comparison table
  - Strategic recommendations
  - Actionable insights

### 🛠️ MCP Architecture
- **FastMCP Server**: Lightweight, high-performance tool hosting
- **Tool Isolation**: Each analysis function is a callable MCP tool
- **Scalable Design**: Easy to add new tools or extend functionality
- **Robust Error Handling**: Graceful fallbacks and error management

## 📋 Prerequisites

- Python 3.8+
- OpenAI API key (get at https://platform.openai.com/api-keys)
- Internet connection (for web search and scraping)

## ⚙️ Installation

1. **Clone the repository**:
   ```bash
   git clone <repository-url>
   cd single-agent-competitive-analysis-agent
   ```

2. **Install dependencies**:
   ```bash
   pip install -r requirements.txt
   ```

3. **Verify installation**:
   ```bash
   python -c "import gradio; import openai; import requests; import beautifulsoup4; print('✓ All dependencies installed')"
   ```

## 🚀 Running the Application

### Quick Start (One Command!)
```bash
python app.py
```

That's it! The application will:
1. Start the Gradio web interface at **http://0.0.0.0:7860**
2. Initialize the analysis engine
3. Open in your browser automatically (or visit http://localhost:7860)

### Expected Output
```
Running on local URL:  http://0.0.0.0:7860
Opening in browser...
```

## 💡 Usage Guide

### Basic Workflow

1. **Enter Company Name**: Type the name of the company you want to analyze
   - Example: "Tesla", "Spotify", "Microsoft", "Stripe"

2. **Provide OpenAI API Key**: Paste your OpenAI API key
   - Get one at: https://platform.openai.com/api-keys

3. **Click "Analyze Competitors"**: The system will:
   - Validate the company exists
   - Identify its sector
   - Find top 3 competitors
   - Analyze competitor strategies
   - Generate comprehensive report

4. **Review Analysis Report**: The report includes:
   - Company overview and sector
   - Top 3 competitors
   - Detailed competitor comparison
   - Strategic insights
   - Actionable recommendations

## 📁 Project Structure

```
single-agent-competitive-analysis-agent/
├── app.py                 # Complete application (Gradio UI + Analysis Engine)
├── requirements.txt       # Python dependencies
├── .env.example          # Environment variables template
├── README.md             # This file
└── ARCHITECTURE.md       # Detailed architecture documentation
```

### Application Components

The `app.py` file includes:

#### 🔍 Analysis Tools
- **validate_company**: Company existence verification via web search
- **identify_sector**: Industry classification with multi-strategy analysis
- **identify_competitors**: Competitor discovery and ranking
- **browse_page**: Web content extraction and parsing
- **generate_report**: Structured competitive analysis report generation

#### 🤖 AI Agent
- OpenAI GPT-4 integration for strategic reasoning
- Orchestrates analysis tools in logical sequence
- Generates insights and actionable recommendations
- Error handling with fallback modes

#### 💻 User Interface
- Gradio-based web interface
- Professional report formatting in Markdown
- Real-time analysis execution
- Input validation and error handling

## 🔒 Security & Privacy

- ✅ **API Keys**: Never stored, used only in current session
- ✅ **Web Data**: Temporary, not cached
- ✅ **No Tracking**: Local processing only

## 🚀 Quick Start

```bash
# 1. Install dependencies
pip install -r requirements.txt

# 2. Run the application
python app.py

# 3. Open browser to http://localhost:7860
# (usually opens automatically)
```

**That's all!** No need for multiple terminals or separate server startup.
# 4. Enter company name and OpenAI API key
# 5. Click "Analyze Competitors"
```

## 🐛 Troubleshooting

**OpenAI API Key invalid**: Check it starts with `sk-` and is active  
**MCP Server not running**: Run `python mcp_server.py` in separate terminal  
**Web search failing**: Check internet connection and try different company name  
**Rate limit errors**: Wait 5 minutes before next analysis

## 📊 Performance

- Analysis Time: 30-60 seconds
- Report Generation: ~10 seconds
- API Calls: 5-8 requests per analysis
- Max Competitors: 3 (quality optimized)

---

**Version**: 1.0.0 (MCP Architecture)  
**Last Updated**: March 2026