Skip to content
Independent guides for QA & test automationRSSEditorial policy
QA Vibes

The workbench

Small tools that run in your browser

No sign-up, no upload, no account. Everything you paste stays on your machine — the checks run as JavaScript in the page you are reading.

Downloads · Bug reports and test cases

QA templates

A bug report template, a GitHub issue form, and a test case CSV, each with a worked example from a real bug or test run.

  • Bug report sections match the grader's scored checks
  • The worked bug report scores 100 in the grader
  • Checkout test cases with the results we observed
Get the templates →

Practice · Test suites

Practice lab

Run your Playwright tests against a demo shop and 10 variants that each hide one bug, then paste the JSON report to see which bugs your tests caught.

  • Every variant is verified: exactly one reference check fails
  • Generates the Playwright config, one project per variant
  • Hints first, answers behind a second click
Open the lab →

Static analysis · Playwright

Playwright test reviewer

Paste a test to find fixed waits, missing awaits, and assertions that can never fail — or compare a test with the version an AI agent healed.

  • 20 checks, each mapped to its eslint-plugin-playwright rule
  • Generates the ESLint config to enforce them in CI
  • Compare mode flags removed assertions and new skips
Open the reviewer →

Rubric · Defect reports

Bug report grader

Score a defect report against ten weighted checks, set severity and priority, and export clean Markdown or Jira markup.

  • Weighted rubric: steps 20, expected vs actual 15, evidence 15
  • Warns on leaked passwords, tokens, and API keys
  • Severity and priority edit the report text directly
Open the grader →

Rubric · User stories

Acceptance criteria grader

Paste a user story and see what a tester will have to ask before they can write a single test — then export the criteria as Gherkin scenarios.

  • Ten weighted checks: criteria 20, outcomes 15, vague wording 15
  • Quotes the words that commit to nothing: fast, properly, gracefully
  • Gherkin export marks TODO where the story never said
Open the grader →

Compliance · Accessibility

Accessibility report summary

Paste the axe JSON your team already produces and see which findings the European Accessibility Act actually requires you to fix, and which are advice.

  • Scope comes from axe-core's own EN-301-549 tags, not our opinion
  • Flags findings on checkout, login, and payment, which the Act names
  • Copies a summary for the ticket, caveats included
Open the summary →

Static analysis · Cypress

Cypress test reviewer

Paste a Cypress spec to find fixed cy.wait() calls, commands chained after actions, async tests, and APIs removed in Cypress 16, or compare it with the version someone “fixed”.

  • Calibrated against eslint-plugin-cypress on the same code
  • Compare mode flags weakened assertions, new waits, and ignored exceptions
  • Flags Cypress.env(), cy.exec(), and .end(), removed in Cypress 16
Open the reviewer →

Test design · Azure Test Plans

Test case CSV for Azure Test Plans

Paste Gherkin or plain steps and get the CSV Azure Test Plans imports, with the required columns, step numbering, and @parameters from a Scenario Outline.

  • Reproduces Microsoft's documented example CSV exactly
  • Flags missing expected results, titles, Area Path, and State
  • Examples table comes out as parameter values to paste
Open the builder →

Pipelines · Azure DevOps

Azure Pipelines test results checker

Paste your Publish Test Results step and the file your runner wrote, and see why the Tests tab stays empty or the pipeline stays green.

  • Catches the missing condition that skips publishing on failures
  • Matches your file against testResultsFiles with minimatch
  • Reads JUnit from Playwright, pytest, and Surefire
Open the checker →

Test design · Single fields

Boundary value generator

Enter what a field accepts, range by range, and get equivalence partitioning, 2-value and 3-value boundary values exactly as the ISTQB syllabus defines them.

  • Gaps in the specification become partitions of their own, with the question to ask
  • Shows which misplaced boundaries each set of values would catch
  • Exact decimals and text lengths; export CSV, Markdown, or pairwise values
Open the generator →

Test design · Business rules

Decision table generator

List the conditions and what the system does, decide every combination, and get the merged table to test — with the combinations nobody decided listed as questions.

  • Merges columns into “–” only where the outcome really doesn't change
  • Combinations that can't happen are deleted, and help others merge
  • One test case per column as CSV, or the table as Markdown
Open the generator →

Test design · Workflows

State transition test generator

Paste a state diagram in Mermaid syntax and get the state table, the coverage items for each ISTQB criterion, and test cases that reach every state and transition.

  • Empty cells in the state table become invalid-transition tests, one per case
  • Flags states no path reaches, and events that go two ways
  • The practice shop example is checked against the real shop
Open the generator →

Test design · Combinations

Pairwise test case generator

Cover every pair (or triple) of parameter values with a fraction of the test cases, with rules, invalid values, and required cases.

  • 4 parameters × 3 values: every pair in 9 cases instead of 81
  • Rules that survive edits and report missing values
  • Export CSV, Markdown, JSON, or a Gherkin Examples table
Open the generator →

Statistics · Release quality

Escaped defect calculator

Turn defects found before and after each release into an escaped defect rate with a 95% range — and find out whether this release is really different from the last one.

  • Wilson range on every rate, because a few dozen defects is a noisy sample
  • Newcombe interval for the change, so “we halved it” gets checked
  • Compares the latest release with its whole history, not just the last one
Open the calculator →

Statistics · Flaky tests

Flaky test rerun calculator

Turn “it failed 3 times in 15” into a failure rate with a 95% range, the number of passing runs that proves a fix, and what your clean runs rule out.

  • Wilson score range, reliable for small run counts
  • Generates the --repeat-each command to run
  • Defaults are real numbers from a flaky test on this site
Open the calculator →