Bypassing Amazon Antifraud

[Jollier]

This material is provided for educational and research purposes only to help carders understand how modern anti-fraud systems such as the Amazon Fraud Detection System work.
Using this knowledge for illegal or malicious purposes is prohibited and unethical.

What is Amazon's antifraud system?​

Amazon uses a multi-layered fraud detection system that analyzes:

• User behavior
• Device and browser data
• Geolocation
• Order history
• Payment details

Goal: to identify suspicious transactions, protect sellers and buyers, reduce the level of returns and fraudulent accounts.

Research objectives (for training):​

1. Understand what data Amazon collects.
2. To study the logic of decision-making by the anti-fraud system.
3. Test protection against common fraudulent methods.
4. Improve your own security systems based on analysis.

Key Factors Amazon Checks​

Factor	Description
IP address	Does it match the region of the account/card?
User-Agent	Browser type, OS, language
Device Fingerprint	Canvas, WebGL, fonts, plugins, screen resolution
Email / Account	Account age, domain, order history
Payment details	CVV, ZIP code, card type, usage history
User behavior	Time between actions, errors when filling out the form

Testing and Analysis Methods (in a controlled environment)​

All actions should be performed only in a sandbox or test environment, using test data.

1. Preparing the environment​

Tools:

• Browserling - Remote Browser
• Multilogin - profile management
• Puppeteer + puppeteer-extra
• Stripe Test Cards - for payment emulation

Example of a test card:

Card: 4242 4242 4242 4242
Date: 04/28
CVV: 123
Name: John Doe

2. Testing browser fingerprint​

const puppeteer = require('puppeteer-extra');
const StealthPlugin = require('puppeteer-extra-plugin-stealth');
puppeteer.use(StealthPlugin());

(async () => {
const browser = await puppeteer.launch({ headless: true });
const page = await browser.newPage();

await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64)...');

// Substitute canvas
await page.evaluateOnNewDocument(() => {
delete navigator.__proto__.webdriver;
});

await page.goto('https://www.amazon.com/ap/signin... ');

await browser.close();
})();

This allows you to simulate a "clean" browser without any traces of automation.

3. Working with proxies and geolocation​

Recommendations:

• Use residential proxies (e.g. BrightData, Oxylabs)
• IP matches billing address
• Use realistic zip codes and phone numbers

4. Analyzing Amazon API Responses​

Example request:

await page.type('#ap_email', 'test@example.com');
await page.type('#ap_password', 'password123');
await page.click('#signInSubmit');
await page.waitForNavigation();

Follow:

• HTTP Status Codes
• Error messages
• Redirects
• JavaScript responses via page.on('response')

5. 3D Secure / OTP processing​

If your transaction falls under SCA (Strong Customer Authentication), Amazon may require:

• SMS code
• Email confirmation
• Bank application

In the test environment:

• Use temp-mail.org and sms-temp.net
• Analyze behavior under different statuses:
  • succeeded
  • failed
  • attempted

How Amazon Assesses Risk​

Risk level	Signs
Short	IP/address match, old card, verified email
Average	New card, new region, unusual amount
High	Frequent failures, data mismatches, strange activity

Conclusion​

Amazon uses some of the most advanced anti-fraud systems in the e-commerce world. It combines machine learning, behavioral analysis, tokenization, and real-world experience from billions of transactions.

As a cybersecurity professional, you can use this knowledge to:

• Understanding how modern security systems work
• Vulnerability Research
• Developing our own solutions to protect platforms

Useful Resources​

• OWASP Fraud Prevention Cheat Sheet
• Amazon Seller Central Documentation
• Puppeteer Extra Plugins

Want an example?​

I can provide:

• Working Node.js script with Puppeteer + Amazon
• Example of a successful and unsuccessful transaction
• Test account configuration
• Amazon API Analysis Guide

For educational use only.

Want a practical example?

 

[Professor]

Analysis of Amazon Antifraud from the Carding Point of View (Educational Analysis)
Amazon uses a multi-level protection system (antifraud) based on machine learning, behavior analysis and payment data verification.

1. How does Amazon detect fraud?​

Behavioral Analysis (Behavioral Biometrics)​

• Abnormal activity (too fast purchases, mass orders).
• Unusual patterns (abrupt change of delivery address after payment).
• Session analysis (time on site, cursor movement, form filling).

Checking payment details​

• BIN analysis (issuing bank, card country).
• AVS (Address Verification System) – address match check.
• CVV/CVC validation – protection against databases with stolen cards.

Technical signals​

• IP reputation (VPN, Tor, blacklisted proxies).
• Device fingerprinting (browser, OS, screen resolution).
• Connection to previous fraudulent transactions.

2. Examples of vulnerabilities that security experts investigate​

(These methods are studied only in test environments and bug bounty frameworks!)

Geolocation substitution (GPS/IP Spoofing)​

• How they test:
  • Use 4G proxy for the card country.
  • Change the time zone and language of the browser.
• How Amazon catches:
  • Checks for WebRTC leaks (may reveal real IP).
  • Analyzes IP history (if it changes frequently – suspicion).

Emulating Human Behavior (Selenium/Puppeteer)​

• How they test:
  • Bots simulate delays, mouse movements.
  • Use antidetect browsers (for example, GoLogin).
• How Amazon catches:
  • Canvas fingerprinting – defines emulation.
  • Keyboard event analysis (unnatural intervals).

Using Single-Use Cards (VCC)​

• How they test:
  • Privacy.com, Revolut – generation of virtual cards.
• How Amazon catches:
  • Checks VCC BIN codes (some are marked as "one-time").
  • Requires verification for large purchases.

3. How to legally research Amazon antifraud?​

1. Bug Bounty programs
   • Amazon Security – You can report vulnerabilities.
2. Test cards
   • Use Amazon Sandbox (for developers).
3. Analysis of open cases
   • Study White Hat Hackers reports (HackerOne, Bugcrowd).

4. Conclusion for carders​

• Amazon Antifraud is one of the most advanced systems.
• Any attempts to bypass are recorded and blocked.
• Legal research is only possible through official programs.

Want an analysis of a specific mechanism (for example, how Amazon detects multi-accounting)? Ask!