Pre-delivery checklist for frontend tasks — run it before declaring any task complete.
Scanned 9/6/2026
Install to Claude Code
npx -y skills add Dev-Toolbelt/dev-team-agents --skill frontend-done-checklist --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Frontend Done Checklist?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/dev-toolbelt-frontend-done-checklist)More formats (shields.io, HTML) on the badges page.
---
name: frontend-done-checklist
description: Pre-delivery checklist for frontend tasks — run it before declaring any task complete.
---
# Frontend Done Checklist
Run through every item before declaring a task complete:
- [ ] Matches the design system (colors, spacing, typography, component patterns) — cross-checked against `docs/development/reuse-guidelines.md` when it exists, not just self-assessed
- [ ] If `ui-ux-designer` produced a Design Review this session, every `[ISSUE]` in it is resolved (not just `[SUGGESTION]`)
- [ ] Responsive — tested at 375px (mobile), 768px (tablet), 1280px+ (desktop)
- [ ] Accessible — keyboard navigable, no missing ARIA labels, sufficient contrast
- [ ] No console errors or warnings
- [ ] No hardcoded strings for international projects (use i18n keys)
- [ ] LCP measured (Lighthouse or DevTools) — target < 2.5 s
- [ ] No duplicate/redundant API calls — checked against `skills/architecture/data-fetching-integrity/SKILL.md` (network panel: no repeated identical requests per action, no re-fetch on every render, no avoidable waterfall)
- [ ] Linters pass
- [ ] No debug artifacts
- [ ] Browser console: no errors; warnings that require disproportionate effort may be skipped but must be noted
- [ ] No `dangerouslySetInnerHTML` / `v-html` with unsanitized content
- [ ] No auth tokens or PII stored in `localStorage` / `sessionStorage`
- [ ] No type errors — type checker passes with no new errors or warnings (where the language supports it)
- [ ] Tests covering the change pass — load `skills/shared/scoped-test-execution/SKILL.md` **before** invoking any test runner and derive the scope from it; the full suite runs only if the user explicitly asked in this session
- [ ] Bundle impact reviewed — no new dependency added without checking its size and necessity
- [ ] Commit message follows project convention — if none is defined, load and follow `skills/shared/conventional-commits/SKILL.md`
- [ ] No Claude attribution in commit messages or PR body — never add "Co-Authored-By: Claude", "🤖 Generated with Claude Code", or any AI/Claude reference; authorship belongs only to the authenticated git user
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!