Setup and teardown are balanced

When a beforeAll or beforeEach hook creates database rows, writes files, or inserts records without a matching afterAll/afterEach to remove them, test data accumulates across runs. Subsequent test runs operate on a progressively more polluted database: unique-constraint violations start appearing, count assertions become wrong, and test order begins to matter. This makes the test suite flaky and eventually unmaintainable. ISO-25010:2011 recoverability requires that test runs return the system to a clean state.

Every beforeAll/beforeEach that creates state needs a corresponding afterAll/afterEach that removes it. Track created entity IDs in the outer scope and delete them in teardown:

describe('UserAPI', () => {
  let userId: string

  beforeAll(async () => {
    const user = await prisma.user.create({ data: { email: 'test@example.com' } })
    userId = user.id
  })

  afterAll(async () => {
    await prisma.user.deleteMany({ where: { id: userId } })
  })

  it('retrieves the user', async () => {
    const found = await prisma.user.findUnique({ where: { id: userId } })
    expect(found?.email).toBe('test@example.com')
  })
})

For simpler isolation, run each test inside a database transaction and roll it back in afterEach — no manual tracking required.