Designing Interaction States

Community

wdm0006

designing-interaction-states

Designs the states screens actually spend their time in — first-run and empty states, loading and skeletons, inline validation and error messages, success feedback, destructive confirmations, and the microcopy for each — so a product feels finished instead of only working on the happy path. Use when building a form, list, dashboard, upload, checkout or settings screen, adding an async action, writing error or empty-state copy, or reviewing a UI that shows blank pages, raw errors or spinners forever.

Overview

Publisher	wdm0006
Repository	`python-skills`
Skill name	`designing-interaction-states`
Stars	111
Forks	17
Bundled files	Instructions only
License	MIT
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
Self-contained
Everything the model needs lives in the instructions — no extra files to sync.
Open source
Published by wdm0006 on GitHub. Read the source before you install it.

Installation

Install the Designing Interaction States AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/wdm0006/python-skills/tree/main/skills/common/interaction-states
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/wdm0006/python-skills.git /tmp/python-skills
mkdir -p .claude/skills
cp -r /tmp/python-skills/skills/common/interaction-states .claude/skills/designing-interaction-states

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Designing Interaction States in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Designing Interaction States on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Designing Interaction States is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Designing Interaction States

Most screens are designed in one state: full of realistic data, with every request succeeding. Users mostly see the others — empty on day one, loading on a slow network, an error on a typo. Those states are where an app feels either finished or neglected.

The state inventory
Empty and first-run states
Loading
Forms and validation
Errors
Success and feedback
Destructive actions
Microcopy rules
Checklist

The state inventory

For every screen, list and design each of these before calling it done:

State	Question it answers
First run / empty	"What is this and what do I do first?"
Loading	"Is it working?"
Partial / some data	"Does it still look intentional with one item?"
Full / overflow	"What happens with 500 items or a very long name?"
Error	"What went wrong, and what do I do now?"
Success	"Did it work?"
Offline / permission denied	"Why can't I do this?"

Seed local data for each and screenshot them; reviewers should see every state.

Empty and first-run states

An empty state is the product's best onboarding surface. Replace "No items." with:

a one-line explanation of what will appear here,
the single primary action that fills it ("Import contacts", "Create your first project"), as a real button,
optionally a secondary path (sample data, docs link).

Distinguish empty because new from empty because filtered ("No results for 'acme' — clear filters") from empty because of an error (that is an error state, not an empty one).

Loading

Under ~300ms: show nothing (a flashing spinner reads as jank).
Content areas: skeletons shaped like the content, so the layout does not jump.
Actions: disable the button and show progress in it ("Saving…"), preventing double submits.
Over ~10s or multi-step: real progress (steps or percent) and the option to leave.
Never an indefinite spinner: every request has a timeout that becomes an error state with a retry.

Forms and validation

Labels above fields, visible always; help text below the label, not in the placeholder.
Mark optional fields ("(optional)") rather than starring required ones, when most are required.
Validate on blur or on submit, not on every keystroke; never show an error before the user has had a chance to type.
On submit with errors: keep everything the user typed, show a summary at the top linking to each field, move focus to the first invalid field.
Error text sits next to its field, says what is wrong and how to fix it, and is linked with aria-describedby (see the building-accessible-interfaces skill).
Use the right input types (email, tel, number, date, autocomplete attributes) so phones show the right keyboard and browsers can autofill.
One primary button per form, at the end, labelled with the outcome ("Create account", not "Submit").

Errors

A good error message has three parts: what happened, why (if known), what to do next.

Bad	Good
"Error 500"	"We couldn't save your changes. Your edits are still here — try again in a moment."
"Invalid input"	"Enter a date in the future."
"Failed to fetch"	"You're offline. We'll retry when you reconnect."

Never show stack traces, raw exception text or status codes to end users; log them with a reference ID and show the ID.
Errors are recoverable where possible: a retry button, preserved input, a link to the place to fix the cause (e.g. billing settings).
Page-level failures get a designed page (404, 500, expired link) with navigation back into the app, not the framework default.

Success and feedback

Confirm every action that changes data: inline ("Saved ✓" next to the button) for small edits, a toast for background actions, a full success screen for milestones (account created, payment complete) that says what happens next.
Toasts auto-dismiss (~5s) except for errors and anything with an action ("Undo").
Optimistic UI is fine for low-risk actions if failures roll back visibly.

Destructive actions

Prefer undo over confirmation for reversible actions ("Contact archived. Undo").
For irreversible ones, confirm with a dialog that names the object and the consequence ("Delete 'Q3 report'? This removes it for everyone and can't be undone.") and a button labelled with the action ("Delete report"), styled as danger — never a generic "OK".
For high-stakes deletes (an account, a workspace), require typing the name.

Microcopy rules

Plain words, sentence case, second person ("Your invoices"), active voice.
Buttons are verbs naming the outcome; links say where they go.
Be specific with numbers and names ("3 contacts imported, 2 skipped — see why").
Same term for the same thing everywhere (pick "workspace" or "team", not both).
No blame ("You entered an invalid…" → "Enter a valid…"), no jokes in errors.

Checklist

Every screen has designed empty, loading, error and success states
No spinner without a timeout and an error path
Forms keep input on error, focus the first invalid field, and use real labels
Error messages say what happened and what to do; no raw exceptions shown
Destructive actions use undo or a specific, danger-styled confirmation
Screenshots of each state attached to the PR

Frequently asked questions

What does the Designing Interaction States AI skill do?

Why use Designing Interaction States on TypingMind?

Because you install it once and use it with any model. Designing Interaction States is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Designing Interaction States in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/wdm0006/python-skills/tree/main/skills/common/interaction-states. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Designing Interaction States?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Designing Interaction States?

As many as you like. As long as a model supports skills, you can use Designing Interaction States with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Designing Interaction States AI skill free?

Yes. It is published on GitHub by wdm0006 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Landing Report

garrytan

CommunityPopular

Read-only queue dashboard for workspace-aware ship. (gstack)

Gstack Openclaw Ceo Review

garrytan

CommunityPopular

Use when asked to review a plan, challenge a proposal, run a CEO review, poke holes in an approach, think bigger about scope, or decide whether to expand or reduce the plan.

Caveman Stats

Show recorded output and cache-read token usage and mode attribution for the current Claude Code session, or locate the host's native usage report. Trigger: /caveman-stats.

Ci Metrics

Query PyTorch CI, GitHub Actions, HUD, Grafana, and infrastructure metrics. Use when users ask about CI duration, job failures, queue times, workflow trends, runner health, dashboard data, or PyTorch…

Designing Interaction States

Overview