Skip to content

TESTING.md

Latest commit

 

History

History
281 lines (220 loc) · 6.15 KB

TESTING.md

File metadata and controls

281 lines (220 loc) · 6.15 KB
Raw

🧪 Testing Your CoreAI Setup

Quick tests to verify everything is working correctly.

1. Health Checks

Test Python Server

curl http://localhost:5001/health

Expected Response:

{
  "status": "ok",
  "message": "CoreAI Multi-Agent System is running",
  "agents": [
    {"name": "Calendar Agent", "status": "idle", ...},
    {"name": "Meeting Agent", "status": "idle", ...},
    ...
  ],
  "total_agents": 6
}

Test Node.js Proxy

curl http://localhost:3001/api/health

Test Frontend

Open browser to http://localhost:3000

2. Test Individual Agents

Test Weather Agent

curl -X POST http://localhost:5001/invoke \
  -H "Content-Type: application/json" \
  -d '{"message": "What is the weather?", "thread_id": "test123"}'

Test Task Agent

curl -X POST http://localhost:5001/invoke \
  -H "Content-Type: application/json" \
  -d '{"message": "Show my tasks", "thread_id": "test123"}'

Test News Agent

curl -X POST http://localhost:5001/invoke \
  -H "Content-Type: application/json" \
  -d '{"message": "What is in the news?", "thread_id": "test123"}'

3. Test Through Frontend

Test Basic Chat

  1. Go to http://localhost:3000
  2. Click on "Chat" in the sidebar
  3. Type: "Hello, what can you do?"
  4. You should get a response listing available agents

Test Specific Agents

Weather:

  • Type: "What's the weather today?"
  • Expected: Weather Agent responds with temperature, conditions, forecast

Tasks:

  • Type: "Show me my tasks"
  • Expected: Task Agent lists current tasks

News:

  • Type: "What's in the news?"
  • Expected: News Agent shows latest headlines

Calendar:

  • Type: "Check my calendar availability"
  • Expected: Calendar Agent shows free time slots

4. Test Multi-Agent Workflow

Type: "Schedule a meeting for tomorrow at 2 PM"

Expected Workflow:

  1. Supervisor identifies: Calendar + Meeting agents needed
  2. Calendar Agent checks availability
  3. Meeting Agent creates Google Meet link
  4. Calendar Agent books the slot
  5. Response includes all steps + meeting link

5. Test Dashboard Features

View Agent Status

  1. Navigate to "Agents" page
  2. Check all 6 agents are shown
  3. Verify status indicators (idle/active/thinking)
  4. Check success rate percentages

View Real-time Activity

  1. Click the bell icon (top right)
  2. Activity feed should appear
  3. Make a request in chat
  4. Activity should update

Test Connections

  1. Go to "Connections" page
  2. Try toggling connections on/off
  3. Status should update

6. Test Settings

  1. Go to "Settings" page
  2. Change your name
  3. Select different AI persona
  4. Click "Save Changes"
  5. Return to Dashboard - greeting should use new name

7. Performance Tests

Response Time Test

time curl -X POST http://localhost:5001/invoke \
  -H "Content-Type: application/json" \
  -d '{"message": "What is the weather?", "thread_id": "test"}'

Should respond in < 2 seconds

Concurrent Requests

Run multiple requests simultaneously:

for i in {1..5}; do
  curl -X POST http://localhost:5001/invoke \
    -H "Content-Type: application/json" \
    -d "{\"message\": \"Test $i\", \"thread_id\": \"test$i\"}" &
done
wait

All should complete successfully

8. Error Handling Tests

Invalid Request

curl -X POST http://localhost:5001/invoke \
  -H "Content-Type: application/json" \
  -d '{"thread_id": "test"}'

Should return error: "Message is required"

Unavailable Agent

Type in chat: "xyzabc123nonsense" Should get general response from Supervisor

9. Browser Console Tests

Open browser DevTools (F12), go to Console:

Test API Connection

fetch('http://localhost:3001/api/health')
  .then(r => r.json())
  .then(d => console.log('Health:', d))

Test Chat API

fetch('http://localhost:3001/api/chat', {
  method: 'POST',
  headers: {'Content-Type': 'application/json'},
  body: JSON.stringify({
    message: 'Test message',
    thread_id: 'browser-test'
  })
}).then(r => r.text()).then(console.log)

10. Logs Check

View Python Logs

# If using start.sh/start.bat
cat logs/python.log    # Unix
type logs\python.log   # Windows

# Or check the console where Python is running

Look for:

  • ✓ "CoreAI Multi-Agent System is running"
  • ✓ No error stack traces
  • ✓ Agent initialization messages

View Node.js Logs

cat logs/nodejs.log    # Unix
type logs\nodejs.log   # Windows

Look for:

View Frontend Logs

cat logs/frontend.log   # Unix
type logs\frontend.log  # Windows

Look for:

✅ Success Checklist

After testing, you should have:

  • All 3 servers running without errors
  • Health endpoints responding
  • Frontend loads and displays dashboard
  • Chat interface accepts and responds to messages
  • At least 3 different agents tested successfully
  • Dashboard showing real-time data
  • Agent status page showing all 6 agents
  • No error messages in browser console
  • No error messages in server logs

🐛 Common Issues

"Port already in use"

# Stop all services
./stop.sh   # or stop.bat on Windows

# Then restart
./start.sh  # or start.bat on Windows

"Module not found" (Python)

cd backend/agentic
source venv/bin/activate   # or venv\Scripts\activate on Windows
pip install -r requirements.txt

"Cannot find module" (Node.js)

cd backend
npm install

cd ../frontend
npm install

Frontend not loading

  • Check if port 3000 is available
  • Check browser console for errors
  • Verify backend servers are running
  • Try clearing Next.js cache: rm -rf frontend/.next

API Key Errors

  • Verify .env file exists in backend/agentic/
  • Check API keys are valid
  • Ensure no extra spaces or quotes in .env

🎉 All Tests Pass?

Congratulations! Your CoreAI system is fully operational.

Next steps:

  • Explore different agent capabilities
  • Try complex multi-step requests
  • Customize the frontend design
  • Add your own agents
  • Integrate real APIs (Google Calendar, Gmail, etc.)

Happy building! 🚀