Skip to main content

Understand Drift Over Time

Read DriftScore as a moving signal rather than a snapshot. Learn what makes the number move, how baselines complement trends, and how to keep history reliable.

Vibgrate Docs

Vibgrate Help

A DriftScore on its own answers "how far behind am I right now?" The more important question for most teams is "is it getting better or worse?" This concept guide explains how to think about drift over time.

Overview

DriftScore (0-100) summarizes how far your dependencies and platform have drifted from current, supported versions. Because the ecosystem keeps releasing, the score moves even when your code does not. Tracking it across many scans turns a static number into a signal you can manage.

What moves the score

  • New upstream releases that widen the gap (drift increases over time even with no code changes).
  • Upgrades you ship that close the gap (the score improves).
  • Large merges or dependency bumps that cause visible steps in the trend.

Two ways to track movement

Trends in Vibgrate Cloud

Push scans regularly and the dashboard plots DriftScore per repository over time:

vg scan --push --repository-name org/my-service

See View Drift Trends on the Dashboard.

Baselines for delta comparison

Locally, a baseline captures scan state so a later scan can be compared against it — useful for fitness functions and CI gates that fail when drift worsens:

vg baseline
vg scan --baseline .vibgrate/baseline.json

Keeping history reliable

Use a consistent --repository-name so the trend stays continuous, and push on a steady cadence so there are no gaps. See Schedule Recurring Drift Reporting.

Related

The DriftScore and Drift Baselines help articles, plus Executive Summaries for Leadership for communicating the trend upward.