Turn a one-off AI chat into a reusable prompt pack for my team
Extracts the prompts, fixes, and checks from an AI chat that worked into a step-by-step workflow and prompt pack colleagues can reuse.
The prompt
Help me turn a task I just completed in an AI chat into a repeatable workflow and prompt pack my team can reuse.
The chat: [PASTE the conversation, or list the prompts I used in order]
The final output I kept: [PASTE or describe it]
What I had to fix or redo along the way: [CORRECTIONS, dead ends, things the AI got wrong]
How often this task comes up and who else does it: [FREQUENCY, ROLES]
Tools and data rules to follow: [APPROVED TOOLS, data that must not be entered]
Step 1 - Diagnose the chat: which prompts did the real work, which turns were only corrections, and what context I added late that should have been given up front.
Step 2 - Create:
1. Workflow - the steps in order, each marked AI or human, with the inputs each step needs and the person who owns the final result.
2. Prompt pack - one consolidated prompt per AI step, with [PLACEHOLDERS] for anything that changes, and the fixes I learned built in so others skip the same dead ends.
3. Checks - what a person must verify after each AI step (facts, numbers, names, tone, confidentiality) before moving on.
4. When not to use it - cases unusual or high-stakes enough to need a different approach or an expert.
5. Quick-start card - the workflow in 6-8 lines for a Teams or Slack post.
6. Improvement log - a simple template for noting what to change after each use.
Write for a colleague who has never seen my chat. Remove names, figures, and confidential details from my example and replace them with placeholders. Do not add steps or facts that were not in my chat; mark anything you had to assume as [CONFIRM].
Highlighted [brackets] are blanks to replace with your own details.
Fill in the blanks and use this promptMore Working with AI prompts
What to fill in
- PASTE the conversation, or list the prompts I used in order
- PASTE or describe it
- CORRECTIONS, dead ends, things the AI got wrong
- FREQUENCY, ROLES
- APPROVED TOOLS, data that must not be entered
- PLACEHOLDERS
- CONFIRM
How this prompt is built
This prompt follows the RTF structure (Role, Task, Format). RTF tells the model who to be, what to do and what shape the answer should take. It is the quickest structure that still gives noticeably better results than a bare request.
Related prompts
- Find where AI can help in my work or businessInterviews you about a process and identifies where AI genuinely helps — and where it doesn't.
- Review and improve a prompt I already wroteAudits your prompt for objective, context, constraints, inputs, output format, quality criteria, and verification, then rewrites it.
- Reverse-engineer the prompt behind an outputWorks backwards from text, code, or an image to the prompt that likely produced it.
- Get better at directing AI, not just using itAssesses how you brief, break down, and review work with AI, then gives targeted practice to manage AI like a capable assistant.