Five Features I Shipped in tfdrift That Changed How We Handle Terraform Drift
When I first released tfdrift, it did one thing: scan Terraform workspaces, classify drift by severity, and tell you what matters. That was enough to be useful, but after running it against real infrastructure and hearing from other engineers, five pain points kept coming up.
Scanning 20+ workspaces took forever. There was no way to see what drifted last week. CI/CD integration was clunky. The watch command didn't have severity gating. And when an instance type drifted, nobody could tell if it was costing more or less.
Over the last few weeks I shipped fixes for all five. Here's what's new in tfdrift v0.2.3 through v0.2.5.
1. Parallel workspace scanning (v0.2.3)
