Prompt Chain Design

You are an AI chain-of-thought prompt engineer. Design a prompt chain for [COMPLEX TASK].

Task: [DESCRIBE THE COMPLEX TASK — something that benefits from breaking into stages]
Final desired output: [WHAT THE CHAIN SHOULD PRODUCE]
AI model: [MODEL]
Context that must carry between steps: [INFORMATION THAT EACH STEP NEEDS FROM PREVIOUS]

Chain design:
Step 1 — [NAME]:
Prompt: [THE EXACT PROMPT]
Input: [WHAT GOES IN]
Output: [WHAT COMES OUT — used as input for step 2]

Step 2 — [NAME]:
Prompt: [THE EXACT PROMPT — includes {{STEP_1_OUTPUT}} placeholder]
Input: [PREVIOUS OUTPUT]
Output: [WHAT COMES OUT]

[Continue for all steps]

Final step output: [THE FINISHED RESULT]

Chain analysis:
- Why each step is separate (rather than one long prompt)
- Where the chain is most fragile (most likely to produce wrong output)
- Validation check between steps 2 and 3 (the most common break point)
- How to test each step independently