Skip to content

michaelst/spendable / .claude/skills/qa

QA a change by driving the running dev app like a real QA engineer - start the server and a real browser session, write a checklist from the agreed requirements and the diff, execute every item with evidence, then go looking for anything that looks off even where the diff did not touch. Reports findings by severity and separates regressions from pre-existing problems. Use after a change is implemented and the gates pass, when the user asks to "QA this", "test it in the app", or "check it actually works".

FileKindSize
SKILL.mdskill md10.8 KB
scripts/example-check.mjsscript2.0 KB
scripts/qa-driver.mjsscript3.5 KB
scripts/qa.shexecutablescript5.8 KB

scripts/example-check.mjs

View on GitHub
// Worked example of one checklist item. Copy into the scratchpad, rewrite, run with
// `qa.sh run <file>`. Keep each file to one or two checklist rows so a failure names itself.
//
// Everything printed here ends up in the QA report, so print evidence (values, paths, problems),
// not narration.
const { openQaSession } = await import(process.env.QA_DRIVER)

const qa = await openQaSession({
  port: Number(process.env.QA_PORT),
  debugPort: Number(process.env.QA_DEBUG_PORT),
  scratchDir: process.env.QA_SCRATCH
})

const fail = (message) => {
  console.log(`FAIL ${message}`)
  process.exitCode = 1
}

// --- check: the budgets list renders and the new-budget form opens ---
await qa.goto('/budgets')

const budgets = await qa.evaluate(`document.querySelectorAll('#budgets li').length`)
console.log(`budgets: ${budgets}`)
if (budgets === 0) fail('no budgets: is this an empty state or a broken query?')

await qa.click(`document.querySelector('#new-budget')`)
// Visibility, not the `hidden` class: show_details() is a JS command that sets an inline display,
// so the class it was rendered with is still there on an open form.
const formOpen = await qa.evaluate(`!!document.querySelector('#details-form')?.checkVisibility()`)
if (!formOpen) fail('clicking New did not open the details form')

// Read the returned path with the Read tool. A screenshot catches what an assertion cannot:
// misalignment, an amount formatted as 1234.5 instead of $1,234.50, text clipped out of its column.
console.log(`shot: ${await qa.shot('budgets-new-form')}`)

// --- check: it survives a narrow viewport ---
await qa.resize(375, 812)
console.log(`shot: ${await qa.shot('budgets-mobile')}`)
const overflows = await qa.evaluate(`document.documentElement.scrollWidth > window.innerWidth + 2`)
if (overflows) fail('page scrolls horizontally at 375px')
await qa.resize(1280, 860)

// Always drain. A check that looks right and logs a 500 or a LiveView crash has still failed.
for (const problem of qa.drainProblems()) console.log(`PROBLEM ${problem.kind}: ${problem.detail}`)

await qa.finish()
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 319kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 319kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 319kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 196kUpdated Claude CodeCodexVerifiedInstall