← ClaudeAtlas

design-reviewlisted

Run a structured design critique against the design file and codebase. Checks visual hierarchy, consistency, responsiveness, accessibility, and aesthetic fidelity. Findings are printed with `DR-n` ids; nothing is written to the repo. Use when user wants a design review, critique, QA pass, polish pass, or mentions "review" after building.
muzalee/claude-atelier · ★ 0 · Code & Development · score 60
Install: claude install-skill muzalee/claude-atelier
This skill runs a structured design review of what has been built, measured against the `## Experience` section of the feature's design file and the aesthetic philosophy named there. > **CRITICAL — look at the running application** > > You MUST view the running application at each breakpoint as part of every design review. Code review alone is insufficient — you need to see what the user sees. Follow the capture protocol in Step 3 below. This is not optional. > > **Nothing is saved.** Take a screenshot, look at it, review from it. No `screenshots/` folder, no image committed to the repo. The findings are the output. **The review is printed, not filed.** Findings carry `DR-n` ids so a fix pass can report against them and a PR can name what is still open. The fix pass records what it did in the design file's `## Implementation` section; this skill writes nothing. ## Example prompts - "Review what I just built" - "Run a design critique on the landing page" - "Check this against the brief" - "Here's a screenshot. How does it look?" [paste screenshot] - "QA pass before I ship this" ## Process 1. **Read the design.** Find the feature's design file by the procedure in `${CLAUDE_SKILL_DIR}/../design/SKILL.md` → **Finding the design file**. Read `## Experience` for the philosophy and the component list, its `### Tokens` block for the token roles, and `## Structure` for the routes to visit. If several features are in play, ask the user which to review. If nothing exists, ask the