regression-and-parity-check

Compare current changes against a baseline to detect regressions.

1|Updated Jan 22, 2026

Search Tags:#validation #migration #quality #regression #baseline #parity

Ready to run

npx skills add https://github.com/hohai99/antigravity-skills --skill regression-and-parity-check

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: regression-and-parity-check
Source: https://github.com/hohai99/antigravity-skills/tree/main/global_skills/regression-and-parity-check
Command: npx skills add https://github.com/hohai99/antigravity-skills --skill regression-and-parity-check

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates parity verification by comparing current changes against a baseline to detect regressions.

Core Features & Use Cases

Parity comparison against a historical baseline to detect deviations.
Classify deviations into Bug, Spec Gap, or Accepted Deviation.
Enforce parity as a gate before release.

Quick Start

Run the regression-and-parity-check skill after implementing a feature or during migration to generate a parity report and regression log.

Frequently Asked Questions about regression-and-parity-check

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I detect regressions during a code migration?▼

To detect regressions during a code migration, compare current changes against a historical baseline. This parity verification process requires a behavior catalog or test suite, access to historical outputs, and a deviation taxonomy to classify differences accurately.

Can I enforce functional parity as a gate before a major release?▼

Yes, you can enforce functional parity as a release gate by comparing current changes against a baseline. This ensures observable behavior is preserved and prevents regressions from reaching production during critical refactors or finalization.

What is the best way to classify deviations found during baseline comparison?▼

The best way to classify deviations during baseline comparison is by using a clear deviation taxonomy. This taxonomy separates detected differences into three specific categories: Bug, Spec Gap, or Accepted Deviation.

Do I need a behavior catalog to verify functional parity?▼

Yes, you need a behavior catalog or test suite to verify functional parity. Access to historical outputs and a clear deviation taxonomy are also required to classify differences and detect regressions accurately.

When should I run a parity check during a critical refactor?▼

You should run a parity check after implementing a feature or during critical refactors. This generates a parity report and regression log, classifying deviations to ensure observable behavior is preserved against the baseline.