🧪TDD Challenge·advanced·⏱️ 40–55m·⭐ 300 XP
M-023Multi-Stage Prompt Pipeline
Description
Nebula Corp's content generation system needs to produce high-quality blog posts through a multi-stage pipeline. Stage 1: Research and outline generation. Stage 2: Write the first draft. Stage 3: Critique and identify improvements. Stage 4: Produce the final polished version. The current system tries to do everything in one prompt and produces inconsistent quality. Build a prompt chaining system where each stage's output feeds into the next, and each stage has a specific, focused responsibility.
Test Cases (10)
Research prompt includes topic
The research prompt must include the topic
Input:"AI ethics", "business leaders", 800
Expected:CONTAINS:AI ethics
Research prompt requests outline
The research prompt should ask for an outline or structure
Input:"Remote work", "managers", 1000
Expected:CONTAINS_ANY:outline,structure,main points,key points
Draft prompt references outline
The draft prompt must reference the outline from the research stage
Input:"Productivity tips", "professionals", 600
Expected:CONTAINS_ANY:outline,OUTLINE,based on
Draft prompt includes target audience
The draft prompt must include the target audience
Input:"Marketing trends", "small business owners", 900
Expected:CONTAINS:small business owners
Critique prompt evaluates multiple dimensions
The critique prompt should evaluate on multiple quality dimensions
Input:"Tech innovation", "investors", 700
Expected:CONTAINS_ANY:clarity,engagement,structure,Clarity,Engagement,Structure
Critique prompt references blog post
The critique prompt must reference the blog post to be critiqued
Input:"Leadership", "executives", 1200
Expected:CONTAINS_ANY:blog post,BLOG POST,content,post
Finalize prompt references original
The finalize prompt must reference the original post
Input:"Data privacy", "consumers", 800
Expected:CONTAINS_ANY:original,Original,previous
Finalize prompt references critique
The finalize prompt must reference the critique
Input:"Sustainability", "general public", 1000
Expected:CONTAINS_ANY:critique,Critique,feedback,weaknesses
Finalize prompt instructs addressing weaknesses
The finalize prompt should instruct addressing identified weaknesses
Input:"Cybersecurity", "IT professionals", 900
Expected:CONTAINS_ANY:address,fix,improve,improvements
All prompts include word count
All prompts should reference the target word count
Input:"Innovation", "entrepreneurs", 750
Expected:CONTAINS:750
Related Lessons
Click Run / Check to validate your solution