Prompt Chain Designer
Splits a complex task into a sequence of simple prompts, with an input, output and check for each step.
The explanation is in the selected language; the prompt text stays in English.
Why use it
For anyone whose single big request gives weak results: split a complex job into steps and keep control of each one.
How to use it
Write the end goal in [GOAL], your material in [DATA OR DOCUMENTS], what good looks like in [WHAT COUNTS AS GOOD] and your tool in [TOOL].
Short example
Goal: turn a long interview into a Telegram post. Material: a 40-minute transcript. Quality: 120 words, plain language. Tool: a chatbot.
What to expect
A numbered chain where each step shows purpose, input, a ready prompt, output format and a check, ending with a plain-text diagram of the chain.
Precautions and tips
- Do not skip the step marked for human review.
- One step does one thing, which makes errors easy to locate.
- Trial the chain on a small sample before using it on the real job.
# Prompt Chain Designer
## Role
You are a workflow designer who builds reliable multi-step prompt chains.
## Context
- Final goal: [GOAL]
- Available inputs: [DATA OR DOCUMENTS]
- Quality bar: [WHAT COUNTS AS GOOD]
- Tool or model I use: [TOOL]
## Task
Decompose the goal into three to six steps. For each step write the purpose, the input it receives, a ready prompt, the expected output format and a quick check that decides whether to continue or retry. Mark which step needs human review and why.
## Output format
A numbered chain. Each step has the labels Purpose, Input, Prompt, Output format, Check. End with a diagram of the chain in plain text.
## Constraints
- Each step does one thing
- Output of one step must fit the input of the next
- Keep every prompt under 150 wordsThe prompt text is the original English and is not translated: paste it into your AI tool as is. Replace [text in square brackets] or CAPITALIZED placeholders with your own details. Always check the answer.
Terms used in this prompt
Interactive mode: search, progress and Python exercises.