Skip to content
Back to search

frontend-toolkit

MarketplaceWorks with Claude CodeGitHub

Claude Code plugin: frontend code-review and browser UI-verification agents, plus a Stop hook that asks for browser evidence after UI edits.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/wiktorsmura--claude-frontend-toolkit/badge.svg)](/listings/wiktorsmura--claude-frontend-toolkit)

HTML

<a href="/listings/wiktorsmura--claude-frontend-toolkit"><img src="/api/listings/wiktorsmura--claude-frontend-toolkit/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

/plugin marketplace add WiktorSmura/claude-frontend-toolkit
/plugin install frontend-toolkit@frontend-toolkit

Restart Claude Code, then check `/hooks` for a `Stop` entry from `frontend-toolkit`.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 13 lines
  1. {
  2. "name": "frontend-toolkit",
  3. "owner": {
  4. "name": "Wiktor Smura"
  5. },
  6. "plugins": [
  7. {
  8. "name": "frontend-toolkit",
  9. "source": "./",
  10. "description": "Frontend code-review and browser UI-verification agents, a Stop hook that asks for browser evidence after UI edits, and React frontend skills."
  11. }
  12. ]
  13. }
README.md · README · 171 lines
  1. <p align="center"><img src=".claude-plugin/icon.png" width="96" alt=""></p>
  2. # frontend-toolkit
  3. A [Claude Code](https://code.claude.com) plugin for React frontends. It gives Claude two read-only
  4. specialists, one for code review and one for UI verification in a real browser, plus a Stop hook
  5. that won't let a UI change end with "it compiles" as the only proof.
  6. | | What | Needs |
  7. | --- | --- | --- |
  8. | **Agent** | `frontend-reviewer`: severity-graded review of a frontend diff | nothing |
  9. | **Agent** | `ui-verifier`: checks the running app against a Figma node or named criteria | a browser MCP; Figma MCP for design-diff |
  10. | **Hook** | `Stop` gate: asks for a browser screenshot after UI files change | bash |
  11. | **Hook** | PR gate (opt-in): blocks `gh pr create` until `frontend-reviewer` has run on the branch | bash, git |
  12. | **Skills** | implementation, architecture, review, a11y, performance and verification guides | nothing |
  13. ## Install
  14. ```
  15. /plugin marketplace add WiktorSmura/claude-frontend-toolkit
  16. /plugin install frontend-toolkit@frontend-toolkit
  17. ```
  18. Restart Claude Code, then check `/hooks` for a `Stop` entry from `frontend-toolkit`.
  19. ## First use (30 seconds)
  20. On a branch with frontend changes, ask:
  21. > Use the frontend-toolkit:frontend-reviewer agent to review my changes against main.
  22. You get findings graded critical / warning / suggestion, each with `file:line` and a concrete fix,
  23. then APPROVE or REQUEST CHANGES. The reviewer runs your repo's typecheck, lint and test commands
  24. instead of re-reviewing what they already catch. It spends its attention on what tools can't see:
  25. logic in the wrong layer, server data copied into client state, missing loading and error states,
  26. effect misuse, accessibility, and drift from your documented conventions.
  27. ## The agents
  28. **`frontend-reviewer`** reads your `CLAUDE.md`, in-repo skills and neighbouring code before judging
  29. anything, and treats violations of *your* conventions as findings. It only reports problems it can
  30. tie to user impact or real maintenance cost, so it won't pad the list with taste.
  31. **`ui-verifier`** has two modes:
  32. - **design-diff**: exports the Figma node, captures the same state in the app, and compares
  33. screenshots *and* computed styles (±2px tolerance, exact token colours). Verdict PASS or FAIL.
  34. - **visual-check**: no design needed. It states the criteria first ("the dialog title doesn't wrap
  35. at 1280px"), then reports each one as MET or NOT MET with evidence.
  36. Every report ends with `Covered:` and `Not verified:` lines, so gaps are stated rather than hidden.
  37. The verifier never starts or kills dev servers and never handles credentials: if it hits a login
  38. wall, it asks you to log in. If your repo has its own browser or dev-server skill, the verifier
  39. lets that skill handle ports, login and feature flags.
  40. ```
  41. Use the frontend-toolkit:ui-verifier agent: design-diff of the settings dialog at
  42. http://localhost:5173 against <figma-node-url>.
  43. ```
  44. ## The Stop hook
  45. When Claude is about to finish a turn, the hook checks the session transcript, subagents included.
  46. If `.tsx/.jsx/.vue/.svelte/.astro/.css/.scss` files were edited after the last browser screenshot,
  47. it blocks the stop once with a request: verify in the browser, or say in one line why no user
  48. could see the change.
  49. - Test, spec and story files don't count, and neither do files under `__tests__/`.
  50. - Any tool with "screenshot" in its name or action counts as evidence: Claude in Chrome,
  51. Playwright, chrome-devtools, Puppeteer, preview tools, or a `screenshot` command run from the
  52. shell. Figma exports don't count, since they show the design, not the app.
  53. - It fires at most once per edit, so it can't trap Claude in a loop.
  54. - It stays off in React Native, Expo and Electron projects.
  55. - Turn it off with `FRONTEND_TOOLKIT_UI_GATE=off`, globally in `~/.claude/settings.json`
  56. (`"env": {"FRONTEND_TOOLKIT_UI_GATE": "off"}`) or for one repo in `.claude/settings.local.json`.
  57. ## The PR gate (opt-in)
  58. Set `FRONTEND_TOOLKIT_PR_GATE=on` (in `~/.claude/settings.json` under `"env"`, or per repo in
  59. `.claude/settings.local.json`) and `gh pr create` is blocked on any branch whose commits change UI
  60. files, until `frontend-reviewer` has run on that branch.
  61. - When the reviewer finishes, a `SubagentStop` hook writes a marker for the branch inside the git
  62. directory. The marker is per worktree, never committed, and survives later commits, so fixing
  63. the findings doesn't need a second review. It records that a review ran, not that it passed.
  64. - It follows a leading `cd <dir> &&` in the command. It's a reminder, not a lock: a
  65. `gh pr create` hidden inside a script isn't seen, and anything it can't work out (no git, no
  66. base branch) lets the command through.
  67. - To skip it once, put `FRONTEND_TOOLKIT_PR_GATE=off` in front of the command.
  68. ## The skills
  69. | Skill | Load it for |
  70. | --- | --- |
  71. | `using-frontend-toolkit` | Starting frontend work from a plan or orchestrator: stack detection, gates, what to put in subagent prompts |
  72. | `implementing-frontend` | Components, design tokens, Figma-to-code, state placement; `references/overlays-and-layering.md` for dialogs, popovers and overflow |
  73. | `frontend-architecture` | Where logic lives: components, hooks, commands, services, stores; the validated API boundary |
  74. | `reviewing-frontend` | Review criteria (preloaded by `frontend-reviewer`) |
  75. | `ensuring-accessibility` | WCAG 2.1 AA: keyboard, focus, ARIA, contrast |
  76. | `optimizing-frontend` | Slow renders, bundle size, Core Web Vitals, memory leaks |
  77. | `ui-verification` | Evidence contract and report format (preloaded by `ui-verifier`) |
  78. The skills are written for React 19 with Vite, Tailwind 4, shadcn/Radix, TanStack Query, Zustand
  79. and react-hook-form + Zod. They apply library-specific advice only when `package.json` shows the
  80. library is there; on another stack, they keep to the parts that don't depend on it.
  81. Name a skill in a prompt or plan task to load it reliably ("use
  82. `frontend-toolkit:implementing-frontend`"). Claude also picks them up from their descriptions, but
  83. much less reliably: see [Design notes](#design-notes).
  84. ## Is it working? Measure it
  85. ```
  86. python3 scripts/usage-report.py --since 2026-01-01
  87. ```
  88. Reads your Claude Code transcripts (`~/.claude/projects`), locally and read-only, and reports:
  89. - how many sessions edited UI files, and how many of those ended with a browser screenshot after
  90. the last edit (the same test the Stop gate applies);
  91. - how often each gate blocked;
  92. - which toolkit agents ran and which skills loaded, per project.
  93. Run it before installing to get a baseline, then a few weeks later. If the evidence rate and the
  94. agent runs don't move, the toolkit isn't earning its keep in your setup. Use `--json` for the raw
  95. numbers.
  96. ## Your repo wins
  97. Your `CLAUDE.md`, `AGENTS.md`, in-repo skills and existing code override every default here. The
  98. skills and agents look for them first and never "correct" a codebase toward the toolkit's
  99. preferences.
  100. ## Design notes
  101. This plugin used to route work to its skills with `SessionStart` and `SubagentStart` notes ("load
  102. skill X when doing Y"). Over a month of real use in a production React codebase, 86 sessions and
  103. 203 subagents received those notes, and **none** loaded a skill because of them. What did work was
  104. naming a skill in a prompt, dispatching an agent, and hooks that check results. So 1.0 drops the
  105. routing and keeps those three. Details are in [CHANGELOG.md](CHANGELOG.md).
  106. ## Development
  107. ```
  108. bash hooks/test-ui-evidence-gate.sh # Stop gate self-check
  109. bash hooks/test-pr-review-gate.sh # PR gate self-check
  110. python3 -m unittest discover -s scripts # usage report self-check
  111. claude plugin validate . # manifest and structure
  112. claude --plugin-dir . # try it without installing
  113. ```
  114. **Evals** (`evals/`, run with `claude plugin eval`) check that review requests reach
  115. `frontend-reviewer`, Figma questions reach `ui-verifier`, backend tasks touch neither, and that
  116. the Stop gate fires on a real edit. The whole suite costs about USD 0.25 per run and runs on your
  117. own Claude credential:
  118. ```
  119. claude plugin eval . --trust-plugin --no-publish --ablation none --scaffold \
  120. --allow-tools Edit --model claude-sonnet-5-5
  121. ```
  122. CI runs the self-checks and validation on every push, which is free on a public repo. It runs the
  123. evals only when the repo has an `ANTHROPIC_API_KEY` or `CLAUDE_CODE_OAUTH_TOKEN`
  124. (`claude setup-token`) secret.
  125. Works on macOS and Linux. The hook needs only bash, grep, sed and sort.
  126. ## License
  127. [MIT](LICENSE). Parts of the skills are adapted from
  128. [TheSoftwareHouse/copilot-collections](https://github.com/TheSoftwareHouse/copilot-collections) (MIT).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.