agentleFS
mia-cxmia-cx/skills/skills/ui-copy/SKILL.md

ui-copy

Skill4 starsChanged 28 days agoCC0-1.0View on GitHub

Use when the user asks to write or fix text the interface shows, such as labels, button names, headings, error messages, empty states, confirmations, placeholders, tooltips, or marketing page copy. Stage 3 of ui-design.

9 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Write only what is needed.

    line 8
  2. Distinctive:

    Keep UI prose to controls, results, and requested copy.

    line 12
  3. Distinctive:

    Never add eyebrows or subheadings.

    line 16
  4. Distinctive:

    Keep necessary labels, prices, terms, instructions, accessible alternatives, and error recovery.

    line 18
  5. Distinctive:

    Every displayed field must have a purpose in this product.

    line 20
  6. Distinctive:

    Write from the user's side of the screen: name what people control ("notifications", not "webhook config").

    line 24

“Common” is measured: kept by at least 6 in 10 of the 9 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

UI copy

Write only what is needed. Every string the interface shows is a design decision, and the response skills apply to it.

Show, don't tell

Keep UI prose to controls, results, and requested copy. Add supplementary information, notices, warnings, caveats, or callouts only when the user explicitly asks; the user decides what warrants attention. For example, omit "16.6 d of data" beside a completion estimate unless requested.

Convey through a visual example what a visual example can convey. Prefer a working product interaction. Delete text that repeats what is already visible.

Never add eyebrows or subheadings. No decorative preheading labels, kickers, overlines, or section numbers above headings, and no taglines or subtitle lines under them. This holds for anything rendered, video frames and motion graphics included: a heading stands alone.

Keep necessary labels, prices, terms, instructions, accessible alternatives, and error recovery. A missing label is not brevity. A sentence earns its place when removing it makes an action or decision harder.

Every displayed field must have a purpose in this product; incidental metadata is not decoration.

Name actions

Write from the user's side of the screen: name what people control ("notifications", not "webhook config"). A control says what happens ("Save changes", not "Submit") and keeps its name through the workflow ("Publish" becomes "Published").

Confirmations name their consequence, so the dialog is answerable without reading its body: "Delete this project?" offers "Delete project" and "Cancel", never "Yes", "No", or "OK". Links name their destination ("Read the billing docs", not "Learn more"). Toggles are labelled for the ON state ("Send read receipts"). One capitalization policy per element type; sentence case is the default.

Error messages identify the problem and the recovery. Empty states offer the first action.

Ship whole templated strings with real pluralization. Word order and plural forms differ per language, so a sentence assembled from fragments around a variable breaks in translation.

Keep it out of the UI

Implementation commentary, founder research, and design rationale live in DESIGN.md and the README, never in customer-facing UI.

The stage is done when every visible string has been re-read as rendered, and each one enables an action, names a result, or supplies explicitly requested copy.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.