home/coding/exercise-2-2-quality-comparison

Exercise 2.2 -- Quality Comparison

GPTClaudeDeepSeek··▲ 1,150 copies·updated 2026-07-14

exercise-2-2-quality-comparison.prompt

# Exercise 2.2 -- Quality Comparison

**Context Engineering Layer: Validation** -- Measure the impact of noise reduction on output quality

## Goal

Re-run the same 3 contract reviews with your optimized ~400-word CLAUDE.md and compare the scores against your Module 1 baseline. This proves (or disproves) that cutting noise improves output.

## Prerequisites

- Completed Exercise 1.2 (you have baseline scores)
- Completed Exercise 2.1 (you have the optimized CLAUDE.md)

## What You Have

- Your `CLAUDE-optimized.md` from Exercise 2.1
- Your `baseline-scores.md` from Exercise 1.2
- The same 3 sample contracts
- `measurement-templates/quality-scorecard.md`
- `measurement-templates/before-after-log.md`

## Your Tasks

### Step 1: Deploy the Optimized CLAUDE.md

Copy your `CLAUDE-optimized.md` to `starter-agent/CLAUDE.md` (back up the original first):

when to use it

Community prompt sourced from the open-source GitHub repo shahabmalikAI5/ai-context-engineering-workshop (no explicit license). A "Exercise 2.2 -- Quality Comparison" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.