content-design-reviewer — Claude Code skill: interface copy review
content-design-reviewer is a Claude Code skill that checks interface labels and guidance for clarity and consistency. It ties findings to a user task and its outcome.
MIT ★ 529 updated 2026-11-17
npx skills add ShuaiMXu/happyhands --skill content-design-reviewer -a claude-codeWhat content-design-reviewer does
The skill reviews product language against a critical task, visible copy, and related error or empty states. It checks whether a label matches its action and preserves factual product claims.
Its output contains evidence, findings, suggested edits, and limitations. Each formal finding needs the frozen skill path and exact rule, the original state and source, an observed failure, and the task consequence with the reasoning that connects them. Missing required evidence is marked needs-evidence.
A small heuristic flags potentially ambiguous labels, but a regex match alone is not a formal finding. Candidates stay unadmitted when no defect is demonstrated. The skill does not invent product capabilities or create a second report alongside the current experience audit.
What it helps with
- spot labels whose meaning is unclear
- check whether an action label matches its result
- review error and empty states against a task
- support copy changes with evidence
Example prompts for the agent
- Review the labels and guidance on the checkout screen.
- Check whether the “Continue” button label matches the state shown after clicking it.
- Review the error and empty states for the critical task and suggest evidence-backed copy changes.
How to install content-design-reviewer
1Skills CLI (Vercel)
npx skills add ShuaiMXu/happyhands --skill content-design-reviewer -a claude-code2Manually: put the files in place
~/.claude/skills/content-design-reviewer/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/content-design-reviewer/
1Skills CLI (Vercel)
npx skills add ShuaiMXu/happyhands --skill content-design-reviewer -a codex2Manually: put the files in place
.agents/skills/content-design-reviewer/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ShuaiMXu/happyhands --skill content-design-reviewer -a cursor2Manually: put the files in place
.cursor/skills/content-design-reviewer/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ShuaiMXu/happyhands skills/product-design/content-design-reviewer2Manually: put the files in place
.github/skills/content-design-reviewer/SKILL.md1Skills CLI (Vercel)
npx skills add ShuaiMXu/happyhands --skill content-design-reviewer -a gemini-cli2Platform CLI
gemini skills install https://github.com/ShuaiMXu/happyhands --path skills/product-design/content-design-reviewerManually: put the files in place
.devin/skills/content-design-reviewer/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about content-design-reviewer
How do I install content-design-reviewer in Claude Code?
Run npx skills add ShuaiMXu/happyhands --skill content-design-reviewer -a claude-code. The skill file is placed at ~/.claude/skills/content-design-reviewer/SKILL.md.
What does content-design-reviewer do?
It reviews interface language, labels, and guidance against a user task and its outcome. The result includes evidence, findings, proposed edits, and limitations.
What evidence does the review require?
It needs a critical task, visible copy, and relevant error and empty states. If required evidence or professional review is missing, the output uses needs-evidence.
Source text
SKILL.md · MIT
--- name: content-design-reviewer description: "Review interface language, labels and guidance for clarity, consistency and task completion." --- # Content Design Reviewer Make product language help users understand choices and consequences. ## Required evidence - critical task - visible copy - relevant error and empty states ## Decisions - Use user language - Check label-action alignment - Preserve factual product claims ## Boundary Do not invent product capabilities or conversion claims. Return structured evidence, findings, proposed updates and limitations. Use `needs-evidence` when required inputs or professional review are missing. Never invent an Experience Score or promote a project observation into a universal rule. ## Execution contract and admission The bounded handler supplies: A small exact-label ambiguity heuristic over captured actions. The professional method still requires: Read the label with its surrounding task and resulting state. Continue, Submit or More may be correct; a regex hit alone must not enter formal findings. Deliver into the current Run's existing experience audit or decision; do not create a competing report. Cite the frozen Skill path and exact rule, original state and source, observed failure, task consequence and the reasoning connecting them. Record preservation/counterevidence. With no demonstrated defect, keep the candidate unadmitted and make no change. `complete` describes this handler's bounded output, not professional review or human acceptance.
© ShuaiMXu · MIT · ShuaiMXu/happyhands