Skip to main content

​
General questions

Screen Answerer is a web-based tool that uses Google’s Gemini AI to automatically detect and answer quiz questions displayed on your screen. It can process both text-based questions and image screenshots to provide instant answers.
Screen Answerer captures screenshots of your screen, sends them to Google’s Gemini AI API for analysis, and displays the detected answers in real-time. It uses computer vision to identify quiz questions and natural language processing to determine the correct answers.
The Screen Answerer application itself is free, but it requires a Google Gemini API key which may have associated costs depending on your usage. Google offers free tier API access with certain quota limits.

​
Supported platforms

Screen Answerer works best with:
  • Google Chrome (recommended)
  • Mozilla Firefox (recommended)
Other browsers may work but are not officially supported. Safari, Edge, and older browsers may have limited functionality, particularly with screen capture features.
Screen Answerer is designed for desktop browsers. Mobile browser support for screen capture APIs is limited, so the monitoring feature may not work on mobile devices. However, you may be able to manually upload screenshots on some mobile browsers.
Screen Answerer is platform-independent and works on:
  • Windows
  • macOS
  • Linux
As long as you’re using a supported browser, the application will function on any operating system.

​
Quiz sites and compatibility

Screen Answerer is designed to work with any quiz site or platform that displays questions visually on your screen. It doesn’t integrate directly with specific quiz platforms but instead analyzes whatever is shown on your screen, including:
  • Online learning platforms
  • Testing websites
  • Educational quiz applications
  • Any text or image-based questions
Yes, Screen Answerer can process multiple choice questions. When analyzing a question with multiple choices, the Gemini AI model will identify the correct answer from the provided options. The application is optimized to return only the correct choice(s).
Yes, Screen Answerer supports both text-based and image-based questions. It can analyze screenshots containing diagrams, charts, images, and other visual content along with the question text.

​
API and costs

Yes, you must provide your own Google Gemini API key. You can obtain a free API key from Google AI Studio. The application stores your API key locally in your browser for security.
Google offers free tier access to the Gemini API with certain quota limits. Costs vary based on:
  • The model you choose (flash-lite is faster and uses less quota)
  • Number of requests
  • Input/output token usage
Check Google’s pricing page for current rates and free tier limits.
Screen Answerer implements several rate limiting measures:
  • Client-side: 5-second window between requests per user
  • Server-side: 100 requests per 15 minutes per IP address
  • Internal quota: 50 API calls per minute (protective measure)
  • Google API limits: Varies based on your API tier
The screen monitoring feature is optimized to stay within these limits, sending requests every 5 seconds by default.
Screen Answerer supports two Gemini models:gemini-2.0-flash-lite (Faster)
  • Optimized for speed and lower quota usage
  • Best for simple, straightforward questions
  • Recommended for most users
gemini-2.0-flash (Balanced)
  • More thorough analysis
  • Better for complex questions
  • Uses more API quota
You can switch between models in Settings > Model.

​
Privacy and security

Yes. Your API key is stored locally in your browser’s localStorage and is never sent to the Screen Answerer server. The application communicates directly with Google’s Gemini API using your key. Screenshots are temporarily stored on the server for processing and are immediately deleted after analysis.
Screenshots are:
  1. Captured in your browser
  2. Sent to the Screen Answerer server
  3. Processed and sent to Google’s Gemini API
  4. Immediately deleted from the server after processing
No screenshots are permanently stored by Screen Answerer.
Your API key is stored in your browser’s localStorage, which is:
  • Stored only on your local machine
  • Not accessible to other websites
  • Persistent across browser sessions
  • Only accessible to the Screen Answerer application
The API key is never sent to the Screen Answerer server or any third party except Google’s Gemini API.
No. All processing happens between your browser, the Screen Answerer server, and Google’s Gemini API. Other users cannot see your questions or answers. Each session is completely independent.

​
Technical questions

Screen capture permissions are required for the monitoring feature to automatically detect quiz questions on your screen. The application uses the browser’s getDisplayMedia() API to capture screenshots at regular intervals. You can revoke this permission at any time.
The screen monitoring feature captures and analyzes screenshots every 5 seconds by default. This interval is optimized to balance responsiveness with API quota usage and rate limiting.
Supported image formats:
  • PNG (.png)
  • JPEG (.jpg, .jpeg)
Maximum file size: 5MB
No. Screen Answerer requires an internet connection to:
  • Communicate with Google’s Gemini API
  • Send screenshots for analysis
  • Receive answers
All AI processing happens in the cloud via the Gemini API.
Screen Answerer attempts to start on port 3000 by default. If port 3000 is already in use by another application, the server automatically tries port 3001, then incrementally higher ports until it finds an available one. Check the server console output to see which port is being used.

​
Usage tips

To improve answer accuracy:
  • Ensure questions are clearly visible and in focus
  • Capture the entire question including all answer choices
  • Use good lighting if photographing physical materials
  • Try the gemini-2.0-flash model for complex questions
  • Make sure text is large enough to read clearly
To minimize API usage and costs:
  • Use the gemini-2.0-flash-lite model (faster and cheaper)
  • Only monitor when actively taking a quiz
  • Stop monitoring when not in use
  • Manually submit questions instead of continuous monitoring
  • Monitor your quota usage in Google AI Studio
If you exceed your API quota:
  1. Wait for your quota to reset (check Google AI Studio for reset time)
  2. Consider upgrading to a paid API tier
  3. Use the application more sparingly
  4. Switch to the flash-lite model to reduce quota consumption

Build docs developers (and LLMs) love

Get started for freeTalk to us