Quick Start Tutorial - rmc0315/firecrawl-ollama-system GitHub Wiki

🚀 Quick Start Tutorial

Get up and running with your first AI-powered website analysis in under 10 minutes!

🎯 What You'll Accomplish

By the end of this tutorial, you'll have:

✅ Installed and configured the system
✅ Analyzed your first website with AI
✅ Generated a professional report with charts
✅ Understood the basics of model selection

Time required: 5-10 minutes

📋 Prerequisites

What You Need

Computer running Windows, macOS, or Linux
Internet connection for setup and web scraping
Python 3.8+ (we'll help you check this)
Basic comfort with command line (we'll guide you!)

Don't worry if you're new to this!

This tutorial is designed for complete beginners. We'll explain everything step by step.

🔧 Step 1: System Setup (3 minutes)

Check Python Version

Open your terminal/command prompt and run:

python --version

Expected output: Python 3.8.x or higher

If you see an error or older version:

Windows: Download from python.org
macOS: brew install python3 or download from python.org
Linux: sudo apt install python3 python3-pip

Download the Project

# Option 1: Download ZIP from GitHub and extract
# Option 2: Clone with Git (if you have it)
git clone https://github.com/rmc0315/firecrawl-ollama-system.git
cd firecrawl-ollama-system

Run Automated Setup

# Windows users
python setup_windows.py

# macOS/Linux users  
python setup.py

What this does:

✅ Installs all required Python packages
✅ Checks your Ollama installation
✅ Tests your system compatibility
✅ Creates launcher scripts

Expected output:

✓ Python: Compatible
✓ Dependencies: Installed  
✓ Ollama: Ready
✓ Setup Complete!

🤖 Step 2: AI Model Setup (2 minutes)

Install Ollama (if not already installed)

Windows:

Download from ollama.com/download/windows
Run the installer
Open Command Prompt

macOS:

# Option 1: Download installer from ollama.com/download/mac
# Option 2: Use Homebrew
brew install ollama

Linux:

curl -fsSL https://ollama.com/install.sh | sh

Start Ollama Service

ollama serve

Keep this terminal window open

Install Your First AI Model

In a new terminal window:

# Fast, efficient model (2GB) - great for beginners
ollama pull llama3.2

# Optional: More capable model (5GB) for detailed analysis
ollama pull qwen3

Verify installation:

ollama list

Expected output:

NAME                    ID              SIZE
llama3.2:latest         a80c4f17acd5    2.0 GB

🔑 Step 3: API Key Setup (1 minute)

Get Your Firecrawl API Key

Visit firecrawl.dev
Sign up for a free account
Copy your API key (starts with fc-)

Don't worry about costs - the free tier includes plenty of scraping for learning!

The system will ask for your key when you first run it - just paste it in and it's saved forever.

🎉 Step 4: Your First Analysis (3 minutes)

Launch the System

# Windows - double-click or run:
launch.bat

# macOS/Linux:
./launch.sh

# Or directly:
python universal_firecrawl_ollama.py

Follow the Interactive Setup

The system will:

Ask for your API key (paste it and press Enter)
Detect your models automatically
Show you the main menu

Expected output:

🔥 Universal Firecrawl + Ollama Integration System
🤖 Your Available Models (1 total):
🚀 Fast & Efficient:
  • llama3.2:latest (1925.8MB) - 3.2B

🎉 System ready!

Run Your First Analysis

Choose Option 1: Single Website Analysis

🎯 Select option (1-7): 1

Enter a website URL:

🌐 Enter website URL: https://example.com

Tip: Start with simple sites like example.com, github.com, or wikipedia.org

Describe what you want to analyze:

📋 Analysis task: What is this website about and who is it for?

Select your AI model:

🤖 Select Model: 1 (llama3.2 - Fast & Efficient)

Watch the magic happen:

🔄 Scraping https://example.com...
✅ Scraped 1,247 characters
🤖 Processing with llama3.2...
✅ Analysis completed!

Save Your First Report

💾 Save Report Options:
1. 📄 Text file (.txt)
2. 📊 CSV file (.csv)  
3. 🌐 HTML file (.html)
4. 📋 PDF file (.pdf)
5. 📁 JSON file (.json)
6. 🎨 HTML with Charts (.html)

Select save format (1-6): 6

Choose option 6 for the most impressive result!

🎊 Congratulations! You Did It!

What Just Happened?

🌐 Web Scraping: Firecrawl extracted clean content from the website
🤖 AI Analysis: Your local Ollama model analyzed the content
📊 Report Generation: Created a professional report with charts
🔒 Privacy Maintained: Everything processed locally on your machine

Check Your Results

Look in the firecrawl_reports/ folder for your new report file. If you chose HTML with charts, open it in your browser to see:

Professional formatting
Interactive charts
Detailed analysis
Timestamp and metadata

🎯 What to Try Next

Easy Next Steps:

1. Try Different Websites

🌐 Good starter websites:
• https://github.com - Tech platform analysis
• https://wikipedia.org/wiki/Artificial_intelligence - Educational content
• https://news.ycombinator.com - Tech news summary
• https://stripe.com - Business model analysis

2. Experiment with Analysis Tasks

📋 Try these analysis prompts:
• "Summarize the main features and benefits"
• "Who is the target audience for this website?"
• "What makes this company different from competitors?"
• "Extract the key technical details mentioned"

3. Explore Different Models (if you installed multiple)

llama3.2: Fast, great for quick summaries
qwen3: More detailed, better for complex analysis
Each model has different strengths!

Advanced Next Steps:

4. Try Model Comparison (Menu Option 2)

Analyze the same website with different AI models
See how each model interprets the content differently
Great for important analyses where you want multiple perspectives

5. Structured Data Extraction (Menu Option 3)

Extract specific information like contact details, pricing, features
Perfect for lead generation or competitive research
Outputs clean JSON data

6. Competitive Analysis (Menu Option 4)

Compare multiple competitor websites
Understand market positioning
Generate comprehensive business intelligence

🤔 Troubleshooting

Common Issues & Quick Fixes

"No module named 'firecrawl'"

# Make sure you're in the right directory and virtual environment
pip install -r requirements.txt

"Ollama connection failed"

# Make sure Ollama is running
ollama serve
# In another terminal, test:
ollama list

"Request timeout" during scraping

Try a simpler website first (like example.com)
Check your internet connection
Some websites are slower to scrape than others

API key issues

Make sure your key starts with fc-
Check you have credits at firecrawl.dev
Use Configuration Settings (Option 6) to update your key

Need More Help?

FAQ Wiki Page - Common questions and detailed solutions
GitHub Discussions - Ask the community
Troubleshooting Guide - Comprehensive problem solving

🎓 Understanding Your Results

Reading the Analysis

Your AI model just:

Read and understood the entire website content
Interpreted your specific question about the site
Provided insights based on its training and the content
Formatted everything into a professional report

Model Capabilities

Context Understanding: Grasps the overall purpose and audience
Information Extraction: Pulls out key facts and details
Pattern Recognition: Identifies common structures and themes
Insight Generation: Makes connections and provides analysis

Why This Is Powerful

Time Saving: Analyzed in seconds what would take you minutes to read
Consistent Quality: Same thorough approach every time
Multiple Perspectives: Different models provide different insights
Professional Output: Ready to share or use in presentations

🌟 Success Tips

Getting Better Results

📝 Write Clear Analysis Tasks

✅ Good: "Analyze the pricing strategy and target market"
✅ Good: "Extract the main features and benefits offered"
✅ Good: "Summarize the company's competitive advantages"

❌ Vague: "Tell me about this website"
❌ Too broad: "Analyze everything"
❌ Unclear: "Is this good?"

🎯 Choose the Right Model

Quick summaries: Use fast models (llama3.2)
Detailed analysis: Use reasoning models (qwen3)
Data extraction: Use coding models (qwen2.5-coder)
When unsure: Try model comparison to see different approaches

🌐 Start with Good Websites

Simple sites work best for learning
Avoid sites requiring login
News sites, company websites, and documentation are great
Complex e-commerce sites might be harder initially

Building Your Skills

Start simple - Basic summaries and overviews
Get specific - Ask targeted questions
Compare approaches - Use multiple models
Save examples - Build a library of successful analyses
Share results - Get feedback from others

🎉 You're Now Ready!

What You've Learned

✅ System installation and configuration
✅ AI model setup and management
✅ Website analysis with local AI
✅ Professional report generation
✅ Troubleshooting common issues

Your Next Journey

Examples Gallery - See what others have created
Use Case Library - Find applications for your field
Advanced Techniques - Level up your analysis skills
Community Showcase - Share your own amazing results

Welcome to the world of local AI-powered web analysis! 🚀

Remember: The community is here to help. Don't hesitate to ask questions, share your results, and help others get started too!