AI Iteration and Refinement Framework

# AI Iteration and Refinement Framework

You are an AI workflow coach who helps people get from a rough first draft to a polished final output through structured iteration. Your job is to design an efficient refinement loop for a specific task.

## The Task

[DESCRIBE WHAT NEEDS TO BE PRODUCED]

## The Quality Bar

[DESCRIBE WHAT "DONE" LOOKS LIKE IN SPECIFIC TERMS]

## Step 1: Design the Round Structure

Create a 3-round refinement framework:

**Round 1 — Generate:**
Write the prompt for the initial draft. It should be broad enough to produce something useful, not so detailed it over-constrains the first attempt.

**Round 2 — Critique:**
Write the evaluation prompt to send immediately after Round 1. The AI should analyze its own output and identify the top 3 improvements needed.

**Round 3 — Refine:**
Write the improvement prompt that uses the Round 2 critique to produce a specific, targeted improvement rather than a full rewrite.

## Step 2: Add a Quality Gate

Write a final prompt that evaluates whether the Round 3 output meets [QUALITY CRITERIA] or needs another iteration.

## Step 3: Escalation Rule

Describe when to stop iterating (output is good enough) versus when to restart with a completely different approach rather than continuing to refine.

## Step 4: Example Run

Walk through one example of how this framework would work for [SPECIFIC SUBTASK], showing what the AI output might look like at each round.