Quality Regression Detection

Regression detection is the bridge between a single run and real maintenance. mgpy should preserve improvements and surface backslides early.

Key points

• CLI: On Windows the examples use the recommended py -3.12 -m <module> ... form (for example py -3.12 -m manifestguard ...). On Linux/macOS this usually maps to python3.12 -m ....
• Baseline comparison is more reliable than gut feeling after larger refactors.
• Regressions can come from coverage, complexity, security or packaging and should be reviewed together.
• A saved good state only matters when the team actually compares against it in daily work.

Recommended mgpy workflow

1. Save a clean reference point before bigger changes.
2. After the rewrite, compare intentionally instead of reading only the new run in isolation.
3. When deltas are negative, decide whether they represent a conscious trade-off or an accidental regression.

Quick start

py -3.12 -m manifestguard baseline --save-label pre-change
py -3.12 -m manifestguard check --extended
py -3.12 -m manifestguard baseline --compare pre-change