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 product designers

Prototype Interaction State Agent

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

Prototype Interaction State Agent: what goes in, what the agent does and what you get

What it does

Prototypes often show only the happy path and leave out what happens on errors, timeouts or empty screens. When a prototype is ready for review, this agent explores each task path and probes transitions that have no defined state, such as invalid input, slow responses, no results or lost connection. For each gap it drafts the missing state with a way to recover. After each addition it tests recovery to confirm the user can get back on track. It reports coverage at the end. The designer decides which additions to keep. Edge case: invalid input in a form leads to a dead end, so the agent drafts an inline error with a retry path.

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, continueYes, continueApprovedNoNo 1 STARTS WHEN Prototype ready for review 2 USES A TOOL Explore task paths 3 USES A TOOL Probe errors, timeouts and empty states 4 CHECKS THE RESULT Does every transition have a defined state? If not: draft the missing state. Back to step 3. 5 USES A TOOL Test recovery after each addition 6 CHECKS THE RESULT Can the user recover and continue? If not: revise the drafted state. Back to step 4. 7 YOU APPROVE Designer decides 8 RESULT State coverage review
Read the steps as a list
  1. Prototype ready for review
  2. Explore task paths
  3. Probe errors, timeouts and empty states
  4. Does every transition have a defined state?If not: draft the missing state. Back to step 3.
  5. Test recovery after each addition
  6. Can the user recover and continue?If not: revise the drafted state. Back to step 4.
  7. Designer decidesThe agent waits here for your OK.
  8. State coverage review

How it decides

It probes transitions without a defined state.

  • Every transition needs a defined state.
  • Each error state must offer a way to recover.
  • The designer makes all design decisions.

Make it yours

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

  • Transition types to probe (errors, timeouts, empty states)
  • Prototype tool in use
  • Coverage report format
  • Who decides on additions

What keeps you in control

It always asks you first

  • Design decisions

Hard limits

  • Drafts only.

It stops when

  • Done: coverage complete.

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 happensA booking prototype with 4 task paths is ready on 30 April. The agent probes 26 transitions. The state check fails on 5, including invalid card input, which leads to a dead end. It drafts an inline error with a retry path. The recovery test fails once: retry clears the whole form. It revises the state to keep entered data. All 26 are covered. The designer approves 4 additions.

More agents for product designers