Quality Gate

Quality gates ensure that mgpy does not only observe quality but also protects decisions. A gate without a clear threshold model only creates noise.

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 ....
• Global thresholds and test-specific overrides belong in a traceable manifestguard.json.
• Expensive checks should only run after the baseline signals pass the default gate.
• A good gate is strict enough for regressions without randomly blocking productive work.

Recommended mgpy workflow

1. Measure the current error and warning levels and derive realistic starting thresholds.
2. Enable the gate early in CI and tighten it step by step later.
3. Whenever rules change, document whether the adjustment reflects product reality or just tooling noise.

Quick start

py -3.12 -m manifestguard check --extended
py -3.12 run_manifestguard.py --ci --report .manifestguard/manifestguard-report.json
py -3.12 -m manifestguard baseline --list