stop-describing-ui skill

A skill that strips explanatory copy from interfaces: nine rules, before and after output, precise verbs for generic ones.

  • 93 ★ on the repository
  • Last commit
Skill

Minimal access

Why this level

  • Text-only instructions for interface copy, the skill holds no executable code
All reasons and checks

artouc/stop-describing-ui

Install

Manual install

npx skills add artouc/stop-describing-ui

The command from the README, the agent is chosen in the prompt.

Access and checks

How we review

Minimal accessThe tool runs nothing and makes no network calls.

  • Text-only instructions for interface copy, the skill holds no executable code

This is third-party code. Review the repository files before installing.

What it does

stop-describing-ui applies when an agent writes, edits or reviews interface text: headings, empty states, button labels, toggle descriptions, helper text, dialogs, banners, permission notices and onboarding. The core idea of the skill: an interface shows what can be done and does not explain itself, since every sentence is something the reader must read before acting. Nine rules: name the action in place of describing the feature; use specific verbs like Connect repository instead of New; keep only the most critical consequence; do not describe backend behavior; do not explain undo when the same control undoes it; do not repeat what the page already says; do not state what does not exist; show the way to enable something on screen; back text with other means. On review the agent goes through every string and presents the result as before and after.

Who it is for. Product designers and interface developers who write microcopy with an agent and want it shorter and sharper.

Good fit when

  • A page empty state retells the feature and has to be replaced by one precise action.
  • A confirmation dialog lists every consequence while the reader needs only the main one.
  • Setting toggle descriptions have grown and the interface needs lightening.
  • A review of all strings on a screen with before and after output is needed.

Not a fit when

  • You need long help articles or documentation: the skill shortens interface text.
  • A user would make a real mistake without an explanation: the rule keeps such sentences.
  • Legal or regulatory text has to stay complete.

Example request

Review the settings screen copy with stop-describing-ui and show a before and after table.

Limitations

The skill is a single `SKILL.md` with rules and before and after examples, with no executable code or third-party dependencies. It sets an editorial stance: shorten text until users begin to err without an explanation. The rules should not be applied blindly, since legal, payment and alarming messages often need full wording. Examples in the repository are in English, and for a Russian interface the principles carry over while the examples have to be adapted. There are no per-agent versions, it installs through `npx skills add`.

How to disable. Delete the skill folder from the agent skills directory or uninstall through `npx skills`. Text edits the agent has already made stay in the project.

FAQ

What does the first rule say?

Name the action in place of describing the feature: the heading, control and label already explain what the screen is for.

How should I replace generic verbs?

Use a specific verb with its object. New becomes Connect repository, and Manage and Configure are replaced by concrete actions.

What about consequences of an action?

Keep the main one. The reader infers the follow-on effects, such as disabled auto-replies obviously stopping reservations too.

What does a review result look like?

The agent goes through every string, applies the rules and shows before and after pairs.

How do I install it?

With `npx skills add artouc/stop-describing-ui`. The rules live in `skills/stop-describing-ui/SKILL.md`.

Discussion

No questions or reviews yet.

Skills

A Codex skill: three distinct visual concepts, screenshot review and cutting excess, so interfaces look less templated.

SkillMinimal access135Repository stars

A UI design skill: art direction per product, a shared DESIGN.md, tokens, and layout and accessibility checks in a browser.

SkillExtended access51Repository stars

A skill that bans the stock AI interface with floating cards, gradients and glass and sets plain sizes, radii and shadows like Linear

SkillMinimal access2.7KRepository stars

A UI design skill: three genuinely different directions side by side, one chosen and refined against real screenshots in agents like Claude Code

SkillExtended access3.2KRepository stars

A skill and CLI for interfaces: build, redesign and review around a product's audience, content and identity

SkillExtended access223Repository stars

Make interfaces feel better

make-interfaces-feel-better

A skill about the small interface details: animations, typography, icons, shadows, radii, optical alignment and hit areas

SkillMinimal access4KRepository stars14Installs via CLI and agents

README in short

The README briefly shows the install command, a link to the rules and before and after images. The main content lives in the SKILL.md file.

README badge

Original SKILL.md

---
name: stop-describing-ui
description: Strip explanatory copy out of user interfaces. Use when writing, editing, or reviewing any UI text, including headings, empty states, button labels, toggle and setting descriptions, helper text, dialogs, banners, permission notices, and onboarding
---

# Stop describing UI

An interface shows what the user can do. It does not explain itself. Every sentence is something the user has to read before acting, so keep a sentence only when the user would make a real mistake without it.

Report a problem