Task 02 Task 02 User Scenario Lab Guide
📥 Lab input
- •
day7_meeting_notes.md(meeting notes)
💾 Lab output
- •
day7_scenario_solution.js
1. Story and lab background
The dev team released the booking detail feature, saying it was perfect. But a beta tester picked a booking, then pressed 'back' to check the expert's rating again before confirming. The booking slot date they'd just entered was all reset to blank, and the angry user closed the app.
From the developer's view, each function that saves booking data worked fine. But when a real user flows through one long scenario (User Journey Flow) — sign-up, expert search, booking a time, payment — info gets lost and the flow breaks like this. In this lab, you use Codex AI to trace the whole feature flow from the start to the end of the scenario in one go, and find and fix the breaks along the path.
2. Learning goals
- See the user's journey (Scenario Flow) from every angle, beyond single feature tests.
- Diagnose how earlier state values are kept or lost across a chain of feature calls.
- Use AI guidance to standardize a must-check list that's easy to miss when checking scenarios.
🛠️ Follow-along lab guide
- Create the lab file: press the download button below and save
day7_meeting_notes.mdin theautomation/folder. - Ask Codex for scenario analysis: enter and send the **Codex request prompt** below in the Codex Client prompt box.
- Write a session-keeping script: attach the sessionStorage-based search restore script from the feedback to the real logic, and check that filter categories reload safely when going back.
Download lab materials
Codex request prompt
Task 02 Prompt
Analyze the step-by-step call order in the provided day7_user_flow.md spec (sign-up ➔ expert search ➔ booking detail ➔ mock payment), diagnose the 2 key sections where info is most likely to be lost as users go through this path, and suggest a state-keeping logic structure as alternative code to prevent it.
4. Example output
Codex analyzes the meeting notes and outputs a normalized data entity mapping and Mermaid ERD script like below.
| File | State caching | Info backed up | Where recovery is guaranteed |
|---|---|---|---|
| day7_scenario_solution.js | Using sessionStorage | Search filter values (Category, Rating, etc.) | Keep filters when returning to the list |
5. AI use points
- Find scenario gaps: finds data-passing bottlenecks and restore points between multiple screen specs that you can't see by looking at one source.
- Map UX threats: AI points out, as an outsider, 'steps that tire users' hidden behind 'technically perfect logic'.