cogikyo/dotfiles/copy
Claude Code and agent skill by cogikyo/dotfiles
- community
Checking that the SKILL.md file is still in the repository.
Use with design when writing, reviewing, or trimming user-facing words on any product surface; covers truthful claims, labels, messages, and word-level clutter.
Install
npx vskill@latest install cogikyo/dotfiles/copySKILL.md
Copy
Read design first for the attention principle, shared test, floor, and project override. Work on words; load hierarchy, color, or motion when the approved scope includes visual emphasis.
The test
- Apply the parent's would the reader miss it? test to each string.
- Shorten surviving text as far as clarity allows, keeping the parent's floor.
- Give each string one job and say each idea once.
- _Readers should not have to separate new information from repetition._
Truth
- Never invent values, stats, testimonials, logos, or claims to fill space.
- Empty is better than deceptive; label illustrative values honestly.
- Check the code path behind an error, confirmation, or consequence before describing it.
Clutter tells
[!INFO] Filling spaceWords need a reader need, even when there is room for them.
_Repeated context makes readers spend time separating new information from what they already know._ Apply the parent's test to these word-level tells:
- Repeated context: text that says what the screen already shows.
- An eyebrow or kicker repeating its heading.
- Helper text or a subtitle restating its label.
- An obvious control's tooltip repeating "Add line".
A · B · Cmetadata orKey: valuerows the screen already implies.- A prefix such as "Sort by:" when the value names itself.
- Filler claims: vague promises that do not explain what the reader can do.
- Replace "Supercharge your workflow" with a specific, supported benefit or remove it.
Surfaces
- Actions: name the result, such as "Save view" rather than "Submit"; use no trailing period.
- Errors: say what happened and what to do next, without apology, blame, or codes the reader cannot act on.
- Confirmations: use only for destructive or hard-to-undo actions; name the object and consequence.
- Toasts and notifications: show only results not already visible.
- Empty states: offer one clear next action or nothing.
- Tooltips: use as a last resort; prefer native
titlefor optional hints and ARIA for accessible names. - Check optional hints are available to the intended input methods.
- Labels: keep one line and use sentence case unless the product has a convention.
- Use one term per concept across the product.
- Keyboard hints: show only non-standard shortcuts.
- CLI/TUI output and logs: lead with the result and print only what the reader acts on.
- Landing and marketing pages: give each section one message; use no filler stats or unsourced social proof.
Procedure
- Find every user-visible string in scope, including loading, empty, error, disabled, and permission-denied states.
- Apply the test to each, keeping the parent's floor.
- Prefer a better label, familiar icon, or layout over explanatory text.
- Keep the accessible name and stay within the approved change scope.
- Align terms with the product's dominant term for each concept.
- Check the rendered result when layout decides whether a label is needed or fits.
Credits: Anthropic frontend-design, Impeccable, and NN/g tooltip guidelines.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingconfig/opencode/skills/design/copy/SKILL.md
Popularity
- Repository stars
- 122
- Repository forks
- 6
- 7-day trend score
- 29
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/cogikyo/dotfiles/copy)