Proxy detection logic challenge

Our 'What is my IP Address' tool generally performs well for standard use cases, but we've encountered a significant hurdle with more advanced proxy detection scenarios. We're observing inconsistent IP resolutions, particularly when clients are behind multiple layers of proxies or specific CDN configurations.

The core issue lies in accurately parsing X-Forwarded-For and other related headers to differentiate between the true client IP and intermediate proxy IPs. Our current logic for proxy detection sometimes fails to correctly interpret complex header chains, which inevitably leads to inaccurate IP reporting for the end-user.

For instance, consider a scenario like this:

  // Example problematic header
  X-Forwarded-For: 192.168.1.100, 10.0.0.5, 203.0.113.45
  // Our tool's current output for client IP:
  Client IP: 192.168.1.100 (Incorrectly identified as client, should be 203.0.113.45 or earlier public IP)

How do you robustly handle X-Forwarded-For parsing and proxy detection for nested proxies or varying CDN behaviors to ensure the most accurate client IP identification? Has anyone tackled this specific challenge effectively?