patrickz
Before you start
A chat assistant, a familiar work task and a fictional source brief.
Why this lesson exists
The original is an AI training playbook for real estate, with each prompt stored as a record alongside an example and a review checklist that includes the industry’s compliance rules. This lab builds a small version for your own job outcome, with search and copy before anything else.
Do the exercise
Set your boundary
Pick one job outcome and store three reviewed prompts as records with a category, tags, a worked example and a review checklist. Add search and one-click copy before any other feature.
Pick one job outcome
Pick one job outcome, such as writing a listing description.
Write three prompt records
Write each prompt as a record: title, job outcome, prompt text with fill-ins, example input and reviewed output. Three is enough.
Add a review checklist
Add a review checklist to every record, including any compliance rules for the industry.
Store records with categories and tags
Store records as structured data with categories and tags.
Add search and copy
Add search and one-click copying.
Next sessions, not today
- Weak and improved examples Show a weak and an improved example beside each prompt.
A prompt to adapt
Replace the bracketed parts with your own details.
Help me build a small searchable prompt library for [one job outcome my team repeats]. First version only: three reviewed prompts stored as records, each with a category, tags, a worked example and a review checklist, plus search and one-click copy. Leave weak-versus-improved examples and every other feature for later. A spreadsheet is fine; if this needs any setup, list it before creating files. Then work in five checkpoints and, after each, tell me how to check it: 1. The job outcome in one sentence. 2. Three prompts written as records: title, job outcome, prompt text with fill-ins, example input and reviewed output. 3. A review checklist on every record. Include a caution for any rules that apply in [your industry], for example Fair Housing rules in property marketing. 4. Records stored as structured data with categories and tags. 5. Search and one-click copy.
When your first version is done, paste it and ask:
Review this [draft from your chosen job task] for unsupported claims, invented details, ambiguous prices or numbers, language that stereotypes or excludes people, and anything that should be verified before publication.
Run this experiment
Give a colleague a new fictional example. Observe whether they can adapt the prompt and catch an intentionally unsupported claim without help.
Check your result
Use evidence from your output. A confident explanation from the AI is not enough.
If it isn’t working
Do not teach a prompt as a compliance guarantee. If the task involves regulated advertising, use current authoritative guidance and a qualified reviewer.
Optional. Progress stays in this browser.
Where this came from
Public project repository ↗. The practice lesson is an adaptation, not a verbatim transcript. About the sources.
Prepared September 2026. Tools and interfaces change; use current official setup instructions. Session lengths are estimates.