Skills
.Work. You
Rest

model-ab-test

Community

Automated A/B testing for model downgrades.

Data & Analytics#subagents#evidence#model-evaluation#A/B#verdict#ab-test
Authorfiatkongen
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill automates the evaluation of model downgrades by running A/B comparisons between the current and proposed models, scoring outputs against predefined criteria, and producing a verdict.

Core Features & Use Cases

  • Parallel AB testing: Dispatch two subagents to compare the current and proposed models in parallel.
  • Binary scoring: Score each criterion strictly as pass/fail and generate a formal verdict.
  • Evidence generation: Write per-test evidence files to document outcomes.
  • Use Case: When a proposed model may degrade performance, run model-ab-test to quantify differences and decide whether to downgrade.

Quick Start

Run model-ab-test with a test-definition file to execute the evaluation and generate results.

Dependency Matrix

Required Modules

None required

Components

references

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: model-ab-test
Download link: https://github.com/fiatkongen/saurun-marketplace/archive/main.zip#model-ab-test

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.