This Chrome extension automatically solves the image–selection captcha that appears on the Easebuzz Merchant Dashboard.
This extension demonstrates that Easebuzz's current captcha implementation is fundamentally flawed. The image-selection captcha can be trivially bypassed using vision AI models like OpenAI's GPT-4 Vision. There is no point in using such captchas as they provide zero security against automated attacks while creating unnecessary friction for legitimate users.
If any Easebuzz engineer is reading this: Please remove this ineffective captcha and either make the user flow easier; without captcha barriers or use a better captcha service.
- Runs only on
https://easebuzz.in/merchant/dashboard. - Clicks the captcha START button inside
#captcha-container. - Waits for the 3×3 image grid (rendered as a
<canvas>). - Captures the canvas as a base-64 image, reads the instruction text (e.g. Choose all the curtains).
- Sends both to OpenAI Vision to receive an array of the 5 grid indices (1–9) that match the instruction.
- Automatically clicks the matched buttons and submits the form to solve the captcha.
- Clone or download this folder.
- In Chrome, open
chrome://extensions, enable Developer mode, click Load unpacked and select this directory. - Click the extension icon → Options to store your OpenAI API key.
- Visit
https://easebuzz.in/merchant/dashboardand trigger the captcha.
Your OpenAI key is stored locally via chrome.storage.local and is never injected into page context.
- Switch models in
background.js(OPENAI_MODEL) if required.