Hire an AI interface state and recovery message writer

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A state list for each screen, with what the person sees and can do
• One new message per failure, saying what happened and one thing to do next
• Empty-screen text that tells a new user what to do first
• A build note per message, covering focus, button label and how the screen changes
• A list of facts still MISSING and any check that did not run

How it does the job well

• Start from the user's task and the content, then the layout. Decoration comes last.
• Design every state: empty, loading, error, long text and mobile.
• Keep one visual system. Reuse spacing, type and colour tokens instead of one-off values.
• Check text contrast (at least 4.5:1 for body text) and keyboard focus as you go.

Mistakes it avoids

• Designing only the happy path at desktop width.
• Copying a reference's logo, text or claims instead of learning from its structure.
• Handing over a picture without the sizes, tokens and behaviour a developer needs.

What a good result looks like

An example, not your task: Support tickets keep starting with "it just says error".

• outputs/step-01-screen-states-designer.md. Each screen's states, what people can do in each, and retry and interrupted-work cases.
• outputs/step-02-plain-error-message-writer.md. A plain message for each failure, ranked by cost, with one next step your product supports.
• outputs/step-03-small-interface-details-designer.md. Focus, button labels and screen changes for each message, written for a developer.
• HAND-BACK SUMMARY. Files saved, sources used, MISSING lines, checks missed and one suggested next step.

How the work is checked

• Every error says what happened and gives one action the product really offers.
• Every state in the list has a message or a MISSING: line.
• No message blames the person, and no message leads with a code.
• Every fact, date and screen name traces to something you were given.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Jakob Nielsen and Milton Glaser would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Screen states designer. Lists the states on each screen you gave it, such as empty, loading, offline, blocked and failed, with what the person can do in each. Hands step 2 every failure and empty state with its cause and recovery options.
• Error message rewriter. Writes one message per failure in the person's words, with no blame and one next step your product supports. Ranks them by cost, so lost money and lost data come first. Hands step 3 the final text.
• Small interface details designer. Turns each message into build notes. It names where focus goes, what the button says and how the message appears and clears. Check before moving on, each note matches its message text.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id write-helpful-error-and-empty-states only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/write-helpful-error-and-empty-states/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• Your screens, as screenshots or pasted text, for each place that can be empty, load, fail or be blocked
• The error text your app shows today, and what causes each one (optional, but it helps)
• What people can really do after a failure, such as retry, undo, sign in again or contact support
• Your voice, such as calm and friendly (optional)
• The code for the screens, if your AI can read it (optional)

Helpful context (optional)

• Support tickets or reviews that quote an error message.
• Your style guide or a few messages you already like.

Tools and access

• A chat AI that reads images and files: It can read screenshots and text exports. In plain chat you paste the screen text instead.
• Any good chat AI: Short rewrites from fixed rules are well within any strong chat AI.
• Your error log: Real counts show which messages are shown most often.

Functions and deliverables

• Interface state and recovery message writer: Share screen states, current text and supported recovery actions. Get a state list, clear messages and button, focus and transition notes for developers.
  • A state list for each screen, with what the person sees and can do
  • One new message per failure, saying what happened and one thing to do next
  • Empty-screen text that tells a new user what to do first
  • A build note per message, covering focus, button label and how the screen changes
  • A list of facts still MISSING and any check that did not run
• Screen states designer: Find the empty, loading, error and offline states your screen is missing, then build each one. Use it as a screen states checklist and builder.
  • A list of states that can happen on that screen
  • Copy and layout for each state
  • Error messages with a clear next step
  • Built versions in your code
  Source: sickn33.
• Error message rewriter: Paste your error messages. Get each one rewritten in plain words, with what went wrong and what to do next, costliest first.
  • Each message rewritten in plain words
  • What went wrong in one sentence and the next step
  • Where the error could be prevented before it shows
  • The messages ranked by the cost of the mistake
• Interface interaction details designer: Find what is missing in the small moments of your screen, like a toggle or loading state. Get a score out of ten and a ranked list of fixes.
  • Trigger, rules, feedback and loop notes for each moment
  • A score out of ten against eight checks
  • A short list of fixes in priority order
  • Ideas for motion and feedback
  Source: Wondel.ai.
• Brand profile builder: Share your website. Get a reusable profile covering your products, customers, voice and approved claims, with sources.
  • A folder of eight plain files, from a start file for your AI to a claims list and a change log
  • A tag on every fact, and every claim marked approved, pending or banned
  • A two-sentence diagnosis of your biggest gap, and a list of what's missing and who can answer it
  • A short slice for your next step, if you name one
• Jakob Nielsen: Review a website, app or form through Jakob Nielsen’s recorded approach. Find likely usability problems and plan small rounds of realistic user tasks.
  • Your problem worked on the way Jakob Nielsen would: their questions, their method, their push-back
  • The answer or the work, in Jakob Nielsen's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Milton Glaser: Review a design message using Milton Glaser’s recorded approach. Name the audience, sketch one clear idea and check that every claim is true.
  • Your problem worked on the way Milton Glaser would: their questions, their method, their push-back
  • The answer or the work, in Milton Glaser's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• Support tickets keep starting with "it just says error".
• You are about to launch a new screen and have only designed the happy path.
• A list, inbox or dashboard is blank on day one and new users do not know what to do first.
• A developer asks you for the exact text for every failed state.

Limits to check before starting

• You only have a list of error messages and want better wording: use the Error message rewriter.
• You want a score and full review of one form or screen: use the Error message rewriter or the Forms and instructions rewriter.
• Your product has no recovery steps yet, such as retry or undo: decide those first, then run this.
• The text has fixed legal or payment provider wording: check with the provider before you change it.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Download the workflow pack and add it to Claude, ChatGPT or Gemini, in a chat that accepts files.
• Attach your screens, or paste the text on each one. Paste your current error messages and say what causes each.
• List what people can really do after each failure. Say so if there is nothing, and do not leave it blank.
• Tell your AI to follow the pack and save each step's file. Check the state list from step 1 before step 2 starts.
• Read each new message against your product, and cut any promise it cannot keep. Then give the build notes to your developer.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/write-helpful-error-and-empty-states for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.