Skip to content

QUICK_REFERENCE.md

Latest commit

 

History

History
193 lines (143 loc) · 4.23 KB

QUICK_REFERENCE.md

File metadata and controls

193 lines (143 loc) · 4.23 KB
Raw

AI Council - Quick Reference Card

🚀 Start the Application

# Backend (if not running)
cd web_app/backend
python main.py

# Frontend
start web_app/frontend/index.html

🔑 API Keys Location

web_app/backend/.env

📍 Important URLs

🧪 Quick Tests

# Test orchestrator
python test_orchestrator.py

# Test web app
python test_web_app.py

# Test API endpoint
python -c "import requests; print(requests.get('http://localhost:8000/api/status').json())"

📝 Configuration Files

File Purpose
config/ai_council.yaml Main configuration
web_app/backend/.env API keys
web_app/backend/requirements.txt Dependencies

🎯 Execution Modes

Mode Speed Cost Quality Use For
Fast ⚡⚡⚡ $ ⭐⭐ Quick queries
Balanced ⚡⚡ $$ ⭐⭐⭐ Most tasks
Best Quality ⚡ $$$ ⭐⭐⭐⭐ Complex reasoning

💰 Cost Estimates

Provider Model Cost/1K tokens
Groq Llama 3 $0.0006
Groq Mixtral $0.0002
Google Gemini $0.0013
OpenAI GPT-3.5 $0.002
OpenAI GPT-4 $0.03

🔧 Common Commands

# Install dependencies
pip install -r web_app/backend/requirements.txt

# Install AI Council
pip install -e .

# Start backend
python web_app/backend/main.py

# Run tests
python test_orchestrator.py
python test_web_app.py

# Check logs
# (View terminal where backend is running)

📊 API Endpoints

# Get status
GET http://localhost:8000/api/status

# Process request
POST http://localhost:8000/api/process
Body: {"query": "Your question", "mode": "balanced"}

# Estimate cost
POST http://localhost:8000/api/estimate
Body: {"query": "Your question", "mode": "fast"}

# Analyze trade-offs
POST http://localhost:8000/api/analyze
Body: {"query": "Your question"}

🐛 Troubleshooting

Issue Solution
Backend won't start pip install -r web_app/backend/requirements.txt
"API key not configured" Check web_app/backend/.env
"401 Unauthorized" Verify API key is correct
Frontend shows "offline" Ensure backend is running on port 8000
Using mock models Check backend logs for "Creating real adapter"

📚 Documentation

File What It Covers
FINAL_SUMMARY.md Complete overview
ORCHESTRATOR_GUIDE.md Orchestrator usage
WEB_APP_SETUP.md 5-minute setup
WEB_APP_COMPLETE.md Full web app guide
REAL_AI_SETUP_COMPLETE.md Real AI configuration

🎨 Customization

# Add new model
1. Edit config/ai_council.yaml
2. Add API key to web_app/backend/.env
3. Restart backend

# Change UI colors
Edit web_app/frontend/index.html
Search for: bg-blue-600, bg-purple-600

# Adjust cost limits
Edit config/ai_council.yaml
Set: max_cost_per_request

🔐 Security Checklist

  • API keys in .env (not in code)
  • .env in .gitignore
  • Never commit API keys
  • Set spending limits in provider dashboards
  • Monitor usage regularly

📞 Get Help

  1. Check documentation files
  2. Review backend terminal logs
  3. Test API keys in provider consoles
  4. Visit http://localhost:8000/docs
  5. Open GitHub issue

✅ Quick Health Check

# 1. Backend running?
curl http://localhost:8000/

# 2. Models available?
curl http://localhost:8000/api/status

# 3. Can process requests?
curl -X POST http://localhost:8000/api/process \
  -H "Content-Type: application/json" \
  -d '{"query":"test","mode":"fast"}'

🎯 Example Queries

"What is quantum computing?"
"Explain recursion with examples"
"Write a Python function for binary search"
"Compare microservices vs monolithic architecture"
"What are the pros and cons of TypeScript?"

🚦 Status Indicators

Color Meaning
🟢 Green Operational
🟡 Yellow Degraded
🔴 Red Offline

Quick Start: python web_app/backend/main.py → Open web_app/frontend/index.html → Start chatting!