ui-component-tests
Skill448 starsChanged 12 days agoApache-2.0View on GitHub
Component tests mount a real React component in a real browser. Use them for small-to-medium uikit components (buttons, fields, list rows, badges, cards). Page-level flows belong in the end-to-end suite, not here. Full reference: packages/uikit/UI_TESTING.md. Read it if anything below is unclear. - Tests are colocated with the component and named Foo.ct.tsx (next to Foo.tsx). The *.ct.tsx suffix keeps them separate from future *.test.ts unit tests. - Import the test API from the shared harness, NOT from @playwright/... directly: import…
In this repository3 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ste-writingSkill
- tk-qa-verify-webSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 22
Import the test API from the shared harness, NOT from @playwright/...
- Very specific: line 25
Every mounted component is auto-wrapped in the styled-components ThemeProvider + GlobalStyle by packages/uikit/playwright/index.tsx.
- Very specific: line 86
Do NOT loosen the maxDiffPixelRatio tolerance to force a flaky screenshot to pass — regenerate the baseline instead.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
UI component tests (Playwright CT) for @tonkeeper/uikit
Component tests mount a real React component in a real browser. Use them for small-to-medium uikit components (buttons, fields, list rows, badges, cards). Page-level flows belong in the end-to-end suite, not here.
Full reference: packages/uikit/UI_TESTING.md. Read it if anything below is unclear.
Conventions (do not deviate)
- Tests are colocated with the component and named
Foo.ct.tsx(next toFoo.tsx). The*.ct.tsxsuffix keeps them separate from future*.test.tsunit tests. - Import the test API from the shared harness, NOT from
@playwright/...directly:import { expect, screenshotEachMode, test } from '<relative>/playwright/test';(fromsrc/components/fields/the path is../../../playwright/test). - Every mounted component is auto-wrapped in the styled-components
ThemeProvider+GlobalStylebypackages/uikit/playwright/index.tsx. Components that useuseTheme()just work. If a component needs more providers (i18n / react-query / router), add them in thatbeforeMounthook — don't wrap inside the test. - Tailwind is loaded too (preflight + utilities + design-token vars) and compiled via
ctViteConfiginplaywright-ct.config.ts, so Tailwind utility classes render in tests.
Pattern
import { Foo } from './Foo';
import { expect, screenshot, test } from '../../../playwright/test';
// Default: ONE screenshot per state. Use this for the vast majority of components.
screenshot('Foo default', () => <Foo />);
screenshot('Foo active', () => <Foo active />);
// Behaviour assertions (real browser):
test('Foo fires onChange', async ({ mount }) => {
let value = '';
const c = await mount(<Foo onChange={v => (value = v)} />);
await c.getByRole('textbox').fill('hi');
expect(value).toBe('hi');
});screenshot(title, render, mode?) renders once (desktop by default). Do not reach for
desktop+mobile by default — most components have no breakpoint-dependent layout. Only when a
component has CSS media queries / responsive layout, use the opt-in decorator
screenshotEachMode(title, render, modes?), which snapshots both desktop (1280×800) and mobile
(390×844) with the matching theme variant.
Commands
cd packages/uikit
yarn test:ct # run all
yarn test:ct:changed # only tests affected by the current diff (also used on PR CI)
yarn test:ct:update # regenerate baselines (Linux only; see below)
yarn test:ct --ui # debug: interactive UI mode
yarn test:ct --debug -g "X" # debug: step through tests matching "X"Debugging flags (--ui, --debug, --headed, -g <title>, npx playwright show-report,
show-trace) are documented in UI_TESTING.md → "Debugging locally".
Prereqs: yarn build:pkg (root) once so @tonkeeper/core dist exists; and
npx playwright install chromium once.
Screenshot baselines — important
- One baseline per screenshot, no platform suffix. Committed next to the component, in a
__screenshots__/folder beside the test (e.g.src/components/fields/__screenshots__/Button.ct.tsx/button-primary.png) — never a central dir. - Screenshots run on Linux only. On macOS/Windows they're auto-skipped (so you can't generate
or clobber a baseline locally); behavioural
test(...)blocks still run. A new screenshot test is RED on CI until its baseline is committed. - To create/update baselines, regenerate on Linux: trigger the Component Tests → "Update
screenshot baselines" GitHub Action (
workflow_dispatch, authoritative — matches CI), or run the pinned Playwright Docker image locally (command inUI_TESTING.md).
Do NOT loosen the maxDiffPixelRatio tolerance to force a flaky screenshot to pass — regenerate the
baseline instead.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

