Changelog
New updates and improvements
December 2025 - Visitor Identity (Server-Issued Visitor ID)
We introduced a server-issued, site-scoped visitorId to reliably link events from the same real device across sessions, without depending on client-side storage.

Reliable Device Linking
Introduced advanced server-side visitor identification to track devices across sessions without persistent client-side storage
Implemented consistent ID generation logic to ensure reliable user recognition across visits
Added intelligent matching capabilities to handle minor environment fluctuations and updates
Enhanced Matching Accuracy
Incorporated robust multi-signal device analysis to achieve higher identification precision
Added network-level intelligence and tie-breaking logic to significantly reduce false positives
Improved identification stability across changing network conditions and browsing contexts
Privacy Mode Resilience
Improved identification performance in incognito and private browsing modes
Enhanced stability for privacy-hardened browser configurations and strict privacy settings
Maintained accuracy even when common anti-fingerprinting measures are active
Strict Signal Validation
Implemented advanced validation logic to prevent false positive matches between similar devices
Added strict consistency checks for high-confidence signals to ensure identity integrity
Enforced rigorous verification standards to distinguish between legitimate users and emulation attempts
October 2025 - 🚧 Closed Beta - Internal Testing Phase
Guardian Stack is currently in closed beta with select partners. This changelog tracks our progress as we refine fraud detection accuracy and expand platform capabilities.

Enhanced Bot Detection Algorithm
Improved detection accuracy for sophisticated headless browsers and automation frameworks
Added support for identifying newer anti-detect browser signatures
Reduced false positives on legitimate browser extensions and accessibility tools
VPN Detection Improvements
Enhanced residential proxy detection across major providers
Better accuracy for legitimate privacy tools vs. fraud-focused VPN usage
Added confidence scoring for VPN detection results
Server SDK Helper Functions
Introduced boolean helper functions for quick fraud decisions (isBot(), isVPN(), isTampering())
Added detailed indicator extraction methods for advanced analysis
Improved error handling and null-safe operations across all helpers
Last updated
Was this helpful?