react-testing skill

React Testing

Comprehensive React testing patterns for behavior-focused component tests, custom hook tests, accessibility assertions, and network-level mocking.

When to Activate

• Writing tests for React components, custom hooks, or pages
• Adding test coverage to legacy untested components
• Migrating from Enzyme or class-component-era patterns to React Testing Library
• Setting up Vitest or Jest for a new React project
• Mocking HTTP requests in tests
• Asserting accessibility violations
• Deciding which tests belong in RTL vs Playwright Component Testing vs full E2E

Core Principle

Test what the user sees and does, not implementation details.

A test should:

• Render the component with the same providers it has in production
• Interact with it via accessible queries (role, label) and userEvent
• Assert visible output and observable side effects (callback fired, request sent)

A test should NOT:

• Inspect component state, props passed to children, or which hooks were called
• Mock React itself or framework hooks
• Assert on the number of renders or DOM structure beyond what affects users

Library Choice

Pick one. Do not run RTL + Vitest AND Playwright CT in the same repo unless you have a clear lane separation.

Query Priority

React Testing Library exposes queries in three tiers — use top-down:

1. Accessible to everyone: getByRole, getByLabelText, getByPlaceholderText, getByText, getByDisplayValue
2. Semantic: getByAltText, getByTitle
3. Test IDs (escape hatch): getByTestId

// Best
screen.getByRole("button", { name: /save/i });

// OK for inputs
screen.getByLabelText("Email");

// Last resort
screen.getByTestId("save-btn");

Variants:

• getBy* — throws if no match
• queryBy* — returns null (use for "assert absence")
• findBy* — async, returns a Promise (use for elements that appear after async work)

User Interaction with userEvent

import userEvent from "@testing-library/user-event";

test("submits the form", async () => {
  const user = userEvent.setup();
  const onSubmit = vi.fn();
  render(<UserForm onSubmit={onSubmit} />);

  await user.type(screen.getByLabelText("Email"), "user@example.com");
  await user.click(screen.getByRole("button", { name: /save/i }));

  expect(onSubmit).toHaveBeenCalledWith({ email: "user@example.com" });
});

• Always await userEvent calls
• Call userEvent.setup() once per test, reuse the returned user
• userEvent simulates a real browser sequence; fireEvent dispatches a single synthetic event — prefer userEvent

Async Patterns

// Element that appears after async work
expect(await screen.findByText("Loaded")).toBeInTheDocument();

// Side effect assertion
await waitFor(() => expect(saveSpy).toHaveBeenCalled());

// Element that should disappear
await waitForElementToBeRemoved(() => screen.queryByText("Loading"));

Never setTimeout + assertion — flaky. Use the matchers above.

Network Mocking with MSW

Mock Service Worker mocks at the network layer. The component, hooks, and fetch library all behave exactly as in production.

Setup

// test/setup.ts
import { setupServer } from "msw/node";
import { http, HttpResponse } from "msw";

export const handlers = [
  http.get("/api/users/:id", ({ params }) =>
    HttpResponse.json({ id: params.id, name: "Alice" }),
  ),
  http.post("/api/users", async ({ request }) => {
    const body = await request.json();
    return HttpResponse.json({ id: "new-id", ...body }, { status: 201 });
  }),
];

export const server = setupServer(...handlers);

beforeAll(() => server.listen({ onUnhandledRequest: "error" }));
afterEach(() => server.resetHandlers());
afterAll(() => server.close());

Configure onUnhandledRequest: "error" so any unmocked request fails the test loudly — silent passes are worse than red.

Per-test override

test("renders error on 500", async () => {
  server.use(
    http.get("/api/users/:id", () => new HttpResponse(null, { status: 500 })),
  );
  render(<UserPage id="1" />);
  expect(await screen.findByText(/something went wrong/i)).toBeInTheDocument();
});

Provider Wrapping

Wrap providers once in a test-utils.tsx:

// test-utils.tsx
import { render, RenderOptions } from "@testing-library/react";
import { QueryClient, QueryClientProvider } from "@tanstack/react-query";

export function renderWithProviders(
  ui: React.ReactElement,
  options?: RenderOptions,
) {
  const queryClient = new QueryClient({
    defaultOptions: { queries: { retry: false } },
  });

  return render(
    <QueryClientProvider client={queryClient}>
      <ThemeProvider theme={lightTheme}>
        <MemoryRouter>{ui}</MemoryRouter>
      </ThemeProvider>
    </QueryClientProvider>,
    options,
  );
}

export * from "@testing-library/react";

Then import { renderWithProviders, screen } from "test-utils" in every test file.

Custom Hook Testing

import { renderHook, act } from "@testing-library/react";

test("useCounter increments and decrements", () => {
  const { result } = renderHook(() => useCounter(0));

  expect(result.current.count).toBe(0);

  act(() => result.current.increment());
  expect(result.current.count).toBe(1);

  act(() => result.current.decrement());
  expect(result.current.count).toBe(0);
});

test("useCounter accepts initial value", () => {
  const { result } = renderHook(() => useCounter(10));
  expect(result.current.count).toBe(10);
});

test("useUser fetches user data", async () => {
  // Instantiate QueryClient ONCE per test outside the wrapper so it survives re-renders.
  // Creating it inside the wrapper closure resets cache state on every render, producing flaky tests.
  const queryClient = new QueryClient({
    defaultOptions: { queries: { retry: false } },
  });
  const wrapper = ({ children }: { children: React.ReactNode }) => (
    <QueryClientProvider client={queryClient}>{children}</QueryClientProvider>
  );

  const { result } = renderHook(() => useUser("1"), { wrapper });

  await waitFor(() => expect(result.current.isSuccess).toBe(true));
  expect(result.current.data).toEqual({ id: "1"

• Wrap state-changing calls in act
• Test through the hook's public API only
• For hooks that use context, pass a wrapper

Accessibility Assertions

import { axe, toHaveNoViolations } from "jest-axe"; // or vitest-axe
expect.extend(toHaveNoViolations);

test("UserCard has no a11y violations", async () => {
  const { container } = render(<UserCard user={mockUser} />);
  expect(await axe(container)).toHaveNoViolations();
});

Run axe in component tests for every interactive component. Catches:

• Missing labels on form inputs
• Invalid ARIA usage
• Poor color contrast (limited — JSDOM has no real CSS engine, so this works for inline styles only; visual contrast belongs in Playwright)
• Missing alt text on images
• Heading order violations

Cross-link: skills/accessibility/SKILL.md for the broader a11y testing playbook.

When NOT to Use Snapshot Tests

Snapshots of rendered output:

• Break on every styling change
• Get rubber-stamped during review
• Test implementation detail (DOM structure), not behavior

Acceptable snapshot uses:

• Pure data serialization functions (formatInvoice(invoice) -> stable string)
• Generated config files (e.g., webpack config output)

For visual regression on components, use Playwright/Cypress screenshots or Percy/Chromatic — actual visual diffs, not DOM strings.

When to Reach for Playwright / Cypress

JSDOM (used by Vitest/Jest) cannot:

• Render real layout (flexbox, grid, viewport queries)
• Run native browser animation, CSS transitions
• Test scrolling behavior, drag-and-drop, paste from clipboard
• Handle iframes, popups, downloads, cross-origin flows
• Run real network in a controlled environment with full DevTools support

For any of those, use Playwright Component Testing (component test in real browser) or full E2E. See e2e-testing skill.

Decision boundary:

• A hook, a presentational component, a form with logic -> RTL
• A component whose layout matters or that uses browser APIs not in JSDOM -> Playwright CT
• A full user flow across multiple pages -> Playwright/Cypress E2E

Coverage Targets

Configure via vitest.config.ts / jest.config.js:

// vitest.config.ts
test: {
  coverage: {
    provider: "v8",
    reporter: ["text", "html", "lcov"],
    thresholds: {
      lines: 80,
      functions: 80,
      branches: 70,
      statements: 80,
    },
  },
}

Anti-Patterns

• container.querySelector("...") — bypasses accessibility queries, lets tests pass when real users would fail
• Asserting on number of renders — implementation detail
• jest.mock("react", ...) — never mock React. Refactor the component instead
• Mocking child components by default — tests the integration, not isolation. Mock only when the child has heavy side effects
• Ignoring act() warnings — they signal real bugs (state update after unmount, missing async wrapping)
• Sharing mutable state across tests — flakes when test order changes
• Tests that pass with it.skip() removed — your test does not actually assert what you think