Skip to main content

​
Get your first answer in 5 minutes

This guide will walk you through setting up Screen Answerer and getting your first quiz answer.
1

Get a Gemini API key

Screen Answerer uses Google’s Gemini AI to process questions. You’ll need a free API key to get started.
  1. Visit Google AI Studio
  2. Sign in with your Google account
  3. Click “Create API Key”
  4. Copy your new API key (starts with AIza)
Keep your API key private. Never share it publicly or commit it to version control.
2

Start the application

If you haven’t installed Screen Answerer yet, see the installation guide.Start the server:
npm start
The application will start on http://localhost:3000 (or the next available port if 3000 is in use).
The server automatically finds an available port if 3000 is already in use.
3

Configure your API key

Open your browser and navigate to http://localhost:3000. You’ll see a welcome screen.
  1. Click the ⚙️ settings icon in the top right corner
  2. Navigate to the “API Key” tab
  3. Paste your Gemini API key into the input field
  4. Click “Save Settings”
The application will validate your key automatically. Once validated, you’ll see a success message.
Your API key is stored in your browser’s local storage and never sent to any external servers except Google’s Gemini API.
4

Answer your first question

Now you’re ready to get answers! You have two options:

​
Option 1: Real-time screen monitoring

  1. Click “Start Monitoring” in the Screen Monitor section
  2. Select the screen or window you want to monitor
  3. When a quiz question appears, Screen Answerer will automatically detect it and display the answer

​
Option 2: Submit a question directly

You can also test with a simple text question:
  1. In the question input area (if using a custom frontend)
  2. Type or paste your quiz question
  3. Click submit to get your answer
Screen monitoring works best in Chrome and Firefox browsers.

​
Understanding the interface

​
Status dashboard

At the top of the application, you’ll see two status cards:
  • Status: Shows whether monitoring is active or inactive
  • Questions processed: Counts how many questions have been detected and answered

​
Answer display

The main answer section shows the most recent answer with formatted markdown support, including:
  • Bold and italic text
  • Lists and numbered items
  • Code snippets
  • Links and references

​
Screen monitor controls

  • Start Monitoring: Begin real-time screen capture and question detection
  • Stop Monitoring: End the monitoring session
  • Monitor preview: Shows the current screen capture being analyzed

​
Choose your AI model

Screen Answerer supports two Gemini models:

Gemini 2.0 Flash Lite

Faster - Best for quick answers and rapid-fire quizzes
  • Lower latency
  • Optimized for speed
  • Default option

Gemini 2.0 Flash

Balanced - Better accuracy for complex questions
  • Higher accuracy
  • Better context understanding
  • Slightly slower
To change models:
  1. Open Settings (⚙️)
  2. Go to the “Model” tab
  3. Use the slider to select your preferred model
  4. Click “Save Settings”

​
Testing the API directly

You can also test the API using curl or any HTTP client: 
curl -X POST http://localhost:3000/process_question_with_key \
  -H "Content-Type: application/json" \
  -d '{
    "question": "What is the capital of France?",
    "apiKey": "YOUR_API_KEY",
    "model": "gemini-2.0-flash-lite"
  }'
Response format: 
{
  "answers": [
    "Paris"
  ]
}

​
Rate limits and quotas

Screen Answerer includes built-in rate limiting to protect against API quota exhaustion:
  • Per-IP rate limit: 100 requests per 15 minutes
  • Request interval: Minimum 5 seconds between screen monitoring requests
  • Global quota: 50 API calls per minute (configurable)
  • Retry logic: Automatic exponential backoff on rate limit errors
The free tier of Google Gemini API has daily and monthly quotas. Monitor your usage in the Google AI Studio console.

​
Troubleshooting

Symptoms: Error message when saving API keySolutions:
  • Verify your API key starts with AIza and is exactly 39 characters
  • Check that you copied the entire key without extra spaces
  • Ensure your API key hasn’t been revoked in Google AI Studio
  • Try generating a new API key
Symptoms: No questions detected during monitoringSolutions:
  • Ensure you’re using Chrome or Firefox (recommended browsers)
  • Grant screen capture permissions when prompted
  • Check that the quiz question is clearly visible in the preview
  • Verify your API key is configured correctly
  • Check the browser console for error messages
Symptoms: “Rate limit exceeded” or “429” errorsSolutions:
  • Wait 5-10 seconds between requests
  • Reduce monitoring frequency
  • Check your Gemini API quota in Google AI Studio
  • Consider upgrading to a paid tier if needed
Symptoms: Port already in use errorSolutions:
  • The application automatically tries the next available port
  • Check the console output for the actual port being used
  • Manually specify a port: PORT=3001 npm start
  • Kill any processes using port 3000: lsof -ti:3000 | xargs kill

​
Next steps

Configuration

Explore advanced settings and customization options

API reference

Learn about all available API endpoints

Build docs developers (and LLMs) love

Get started for freeTalk to us