agentleFS
TresnandaTresnanda/kybern/.cursor/skills/better-writing/SKILL.md

better-writing

Skill83 starsChanged 2 months agoMITView on GitHub

Focuses on improving product copy in your project.

25 other projects use a version of this, 17 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use improve-ui, good-css and 5 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Discriminative-shoji607Skill

    improve-ui

    Audit an existing product surface against its own design evidence, identify verified UI problems, and write self-contained implementation plans for another agent. Strictly read-only on product source. Use when asked to review, refine, improve, or clean up an interface without replacing its identity; investigate design-system drift; or prepare a design handoff.

    Also in 3 projects that use this

  • sanmidableSkill

    good-css

    Also in 3 projects that use this

  • agentset-aiSkill

    fixing-metadata

    Ship correct, complete metadata.

    Also in 3 projects that use this

  • VercelSkill

    baseline-ui

    Validates animation durations, enforces typography scale, checks component accessibility, and prevents layout anti-patterns in Tailwind CSS projects. Use when building UI components, reviewing CSS utilities, styling React views, or enforcing design consistency.

    Also in 3 projects that use this

  • studio-lgtmSkill

    make-interfaces-feel-better

    Apply design engineering best practices to polish UI components - use when reviewing or improving interfaces, animations, typography, or interaction details.

    Also in 4 projects that use this

  • swyxioSkill

    mobile-native

    Apply swyx's native mobile UX preferences when designing or critiquing iPhone/iPad flows, adapting tablet layouts, choosing communication-app notification behavior, or opening notifications/deep links/widgets/shortcuts before local content is ready. Also use for iOS/iPadOS generated build configuration, signing/install/launch problems, and device playtesting. Keep existing native or cross-platform stacks; excludes browser-only responsive work and backend-only changes without native UX or delivery implications.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 19 edited versions we read keep these.

  • 3

    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. Distinctive:

    Before writing or reviewing, read the copy nearby.

    line 14
  2. Distinctive:

    Use possessives sparingly: "Favorites" beats "Your Favorites".

    line 33
  3. Very specific:

    Never assemble a sentence from fragments around a variable ("You have " + n + " new messages"), because word order changes per language.

    line 41
  4. Distinctive:

    No blame, no "oops", no exclamation marks.

    line 79

“Common” is measured: kept by at least 6 in 10 of the 19 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

Interface writing

Clear and brief beats clever; consistent beats varied. The best error message is the interaction redesigned so the error cannot happen.

How copy renders (capitalization via text-transform, truncation, smart punctuation) belongs to better-typography. Error markup and announcements (aria-invalid, live regions) belong to better-accessibility. Room for translated strings belongs to better-layout.

Recon the existing voice

Before writing or reviewing, read the copy nearby. Note the product's terminology, its localization conventions and any voice or content style guide.

A deliberate brand voice is not a defect. Raise a departure from plain language only when it creates inconsistency, ambiguity, translation risk, or a tone the stakes don't support.

One voice, flexible tone

The product has one voice and its existing copy establishes it. A local edit does not get to invent a new one. Keep terms consistent: if it's "Archive" in the menu, it isn't "Move to storage" in the toast. Tone flexes with the stakes:

ContextTone
Success, onboarding, empty statesWarm, can be light
Routine actions, settingsNeutral, minimal
Errors, destructive confirmationsCalm, plain, zero playfulness
Data loss, securitySerious, explicit

Address the reader directly

In instructional copy, write "you", not "the user". In errors, "we" invites ambiguity and reads as deflection, so prefer "Unable to load content" over "We're having trouble loading this content". An established first-person voice can stay in low-stakes copy where it still reads clearly.

Use possessives sparingly: "Favorites" beats "Your Favorites". Hold one perspective throughout a flow.

Plain words over clever ones

Choose words a tired reader gets on the first pass, and delete every word that does no work. No idioms, no colloquialisms, no humor that won't translate.

Skip unnecessary gender: "Subscribers can post recipes", not "each subscriber can post his or her recipes". Match the input device: "tap" on touch, "click" with a pointer, "select" when both are possible.

Never assemble a sentence from fragments around a variable ("You have " + n + " new messages"), because word order changes per language. Use a full templated string with proper pluralization.

Verb-first buttons

A button label starts with a verb naming the action: "Send", "Save draft", "Delete project". Never "OK!", "Let's go!", or a bare "Yes" and "No" on a consequential action.

A confirmation button repeats the consequence, so the dialog is answerable without reading the body. "Delete this project?" offers Delete project and Cancel.

Consistent flow vocabulary

A multi-step flow uses one vocabulary throughout: "Get started" to enter, "Continue" or "Next" (pick one) to advance, "Done" to finish. Alternating synonyms makes users wonder whether the buttons do different things.

Links describe their destination

Link text has to make sense out of context, because screen-reader users navigate by a list of the page's links. Write "Read the billing docs". "Click here" fails this and the device-verb rule at once.

A bare "Learn more" breaks down as soon as two appear on one page. Suffix each one: "Learn more about exports".

One capitalization policy

Pick title case or sentence case per element type, then apply it to every instance of that type. Sentence case is the safer default. It is calmer, has no per-word rules to remember and localizes cleanly. "Save Changes" beside "Discard changes" reads as sloppiness.

Settings describe the ON state

Label a toggle for what happens when it is on. "Send read receipts" lets users infer the off state; the negative ("Don't send read receipts") turns the toggle into a double negative.

Link straight to a referenced setting rather than describing the path to it: a "Notification settings" link, not "Go to Settings > Notifications > Email".

Errors say how to fix, next to where it broke

An error is an instruction, and it belongs beside the field that failed:

BadGood
That password is too shortChoose a password with at least 8 characters
Invalid nameUse only letters for your name
Oops! Something went wrong.Unable to save. Check your connection and try again.

No blame, no "oops", no exclamation marks. Phrase hints positively ("Use only letters", not "Don't use numbers or symbols") and show them before the mistake, not after. When the same error keeps firing, redesign the interaction instead of rewording it.

Empty states point forward

An empty state says what this place is, how to fill it and offers one clear next action:

<!-- Bad: a shrug -->
<p>No results.</p>

<!-- Good: orientation plus a next step -->
<p class="font-medium">No projects yet</p>
<p class="text-sm text-zinc-500">Projects keep your tasks and files together.</p>
<button class="mt-4">Create a project</button>

A search or filter empty state names the query and offers an exit: "No results for 'quarterly'. Clear filters". Never park persistent information in an empty state. It disappears the moment content exists.

Placeholders are examples, not labels

A placeholder shows the expected format: name@example.com, DD/MM/YYYY. It vanishes on input, so it is never the only label. Every field keeps a visible one.

Reporting

Severity. HIGH misleads the user or hides how to recover from an error. MEDIUM breaks voice, terminology, or capitalization consistency. LOW is isolated wording polish.

Verification. Source alone is enough here. Check every label against the action it invokes, every error for a stated fix and terminology against the copy around it. No browser check is required.

Format. Group findings under the principle each violates, ordered by severity, one row per root cause listing every location it appears in:

SeverityLocationBeforeAfterWhy

Location is path/to/file:line. Why names the principle and the user impact.

End with Block when any HIGH remains, Approve otherwise, leaving the rest in the table as work to do. Never Approve coverage you did not inspect. With nothing to report, state "No actionable writing findings" and report verification.

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.