Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for user experience designers

User Test Prototype Iteration Agent

Turn test observations into ranked problems and verified changes for the next prototype.

AnalysisDecisionsCreativesProduct development
See how it worksGet access
User Test Prototype Iteration Agent: what goes in, what the agent does and what you get

What it does

After testing a prototype with eight people, the team remembers the loudest comment and fixes that, while the real problem, a latch that five people struggled with, is ignored. This agent records observations by task for each participant, such as success, time, errors and quotes. It groups failures across participants and ranks them by severity and frequency. For the top issues it proposes design changes and checks each one against constraints, such as cost, material, size and safety. It checks that every top issue has at least two pieces of evidence and a proposed change, and sends weak ones back for more evidence. It plans a retest with the same tasks. After the retest, it compares results and confirms the issue is fixed. The designer approves the changes. Edge case: a fix conflicts with a safety rule, so the agent rejects it.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedYes, continueNoNo 1 STARTS WHEN User test session ends 2 USES A TOOL Record observations by task and participant 3 DOES Group failures and rank by severity and frequency 4 CHECKS THE RESULT Does each top issue have evidence from at least twoparticipants? If not: mark it as a possible issue and plan extraobservation. Back to step 3. 5 DOES Propose design changes for the top issues 6 USES A TOOL Check each change against cost, size and safetyconstraints 7 DOES Plan the retest with the same tasks 8 YOU APPROVE Designer approves the changes and the retest plan 9 USES A TOOL Record results of the retest 10 CHECKS THE RESULT Did the failures on the top issues drop to thetarget? If not: propose a different change and plan anotherretest. Back to step 5. 11 RESULT Iteration report
Read the steps as a list
  1. User test session ends
  2. Record observations by task and participant
  3. Group failures and rank by severity and frequency
  4. Does each top issue have evidence from at least two participants?If not: mark it as a possible issue and plan extra observation. Back to step 3.
  5. Propose design changes for the top issues
  6. Check each change against cost, size and safety constraints
  7. Plan the retest with the same tasks
  8. Designer approves the changes and the retest planThe agent waits here for your OK.
  9. Record results of the retest
  10. Did the failures on the top issues drop to the target?If not: propose a different change and plan another retest. Back to step 5.
  11. Iteration report

How it decides

It ranks problems by how many people failed and how badly. A change is considered only if it fits cost, size and safety constraints.

  • Rank an issue higher when 3 or more participants fail the same task
  • Treat an issue with one witness as possible, not confirmed
  • Reject any change that breaks a safety or cost constraint
  • Set a retest target of at most one failure in the group

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Tasks to record
  • Severity scale
  • Constraints list
  • Retest group size (default 6)
  • Target failure level

What keeps you in control

It always asks you first

  • Designer approves the changes and the retest plan

Hard limits

  • Never approve a change that breaks safety rules
  • Never treat a single comment as a finding

It stops when

  • Done: top issues are fixed in the retest
  • Stop: no change fits the constraints, so the designer decides on a trade-off

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensOf 8 participants, 5 fail to open the latch and 2 misread the label. The agent ranks the latch first. It proposes a larger tab, but the check shows it exceeds the size limit by 2 mm. It proposes a redesigned angle instead, which fits. The retest with 6 people shows 1 failure. The label issue stays at 2 failures of 8, so the agent proposes a clearer icon for the next round.

More agents for user experience designers

Compliance · Automation

Prototype Usability Session Agent

Consistent, consent-respecting usability session records.

9 steps2 checks3 jobs
Analysis · Product development

Product Onboarding Friction Agent

Onboarding friction traced to its real cause.

8 steps1 check4 jobs
Creating · Quality control

Prototype Interaction State Agent

Prototypes that cover errors and recovery, not just the happy path.

8 steps2 checks3 jobs
Document handling · Automation

Design Decision Log and Rationale Agent

Keep a decision log where every design choice has an owner, options and linked evidence.

10 steps2 checks3 jobs