Skip to main content

How It Works

Bot Detection​

Absinthe Protect's bot detection system is designed to safeguard your Web3 application from automated agents, ensuring that real users are engaging with your service.

Our bot detection technology scrutinizes a myriad of behavioral and environmental signals to determine the authenticity of a user session.

Signals Examples:​

  • Automated Browser Detection: We check for telltale signs of common automation tools such as Selenium, Puppeteer, or Playwright which might indicate scripted, non-human interaction.
  • IP Analysis: Our system assesses IP addresses to identify any that are blacklisted or have previously been associated with suspicious activities.
  • Viewport Metrics: By analyzing viewport sizes, we can detect unusual patterns that deviate from typical user behavior.
  • Device Fingerprinting: We collect non-invasive device characteristics to recognize and flag potential bot signatures.
  • Virtual Machine Detection: We check for signals if the browser is being executed in a virtual machine, which can indicate a botted environment.

Multi-Accounting​

Multi-accounting refers to the practice where a single user operates multiple accounts, often to exploit application features such as rewards, voting, or airdrops. This can give one user an unfair advantage and skew the application's intended use.

Detection Mechanisms​

Absinthe Protect employs sophisticated techniques to recognize multiple accounts operated by a single user.

  • Browser Fingerprinting: We analyze the unique fingerprint of each user's browser, which encompasses a blend of settings and extensions that are often distinct for individual users.
  • Wallet Association: By observing the interaction patterns of wallets with your application, our system can detect when multiple wallets are likely being controlled by the same entity.

What If They Change Devices?​

A user will still be identified even if they switch devices. Even if a malicious attacker switches to a different browser and changes their IP (without the use of a VPN), they will still be unable to engage in large-scale Sybil attacks.