QA Engineer · Runs on ChatGPT GPT-5.5 · 5 cr./message

AI assistant: finds where it will break

Test cases, boundary conditions, regression checklists and turning a defect into a reproducible step.

What QA Engineer returns

The shape of the answer is fixed in advance: happy path, range boundaries, empty and maximum values, wrong types, repeats and races. Two conversations about the same task therefore end in the same form: the answer can be compared with the previous one and passed on without asking for a rewrite.

“Build checks in groups: happy path, range boundaries, empty and maximum values, wrong types, repeats and races, external service failure, permissions, locale and time zone.”

“Turn a defect into a reproducible scenario: minimal steps, environment, data, frequency.”

What QA Engineer will not do

The role has 7 rules, 2 prohibitions among them outright. We quote them as written rather than paraphrase: promising “it can do anything” is cheaper, but then people meet the limit only after paying for a message. Here it is visible beforehand, and a neighbouring role can be picked straight away.

“Never claim something works or is broken without verification — separate verified facts from assumptions.”

“Do not invent framework or API behavior, version numbers or documentation links.”

How QA Engineer runs the review

The rules describe the role as “QA engineer”. Below is how it holds the review, in the same words it is given before your first message — and manner is exactly what separates one companion from another.

“Write each case as steps, expected result and actual result.”

What a message to the “Finds where it will break” role costs

You are charged per message sent, not per month of access: a role needed once a quarter costs nothing in between, and it needs no line of its own in the budget.

5 credits per messageThe price is the same for a short and a long message: what counts is the send, not the length.No subscription feeTen messages a month cost the price of ten messages: you pay for what happened, not for a plan.

What QA Engineer flags for checking

The danger is not a gap in knowledge but a confident tone over a guess: people act on such answers, and the cost of the mistake lands on the person, not the program. So the rules say separately what to flag, what to warn about and when to ask for a source.

“Start from what the feature promises, to whom and under which conditions: without a promise there is nothing to verify.”

What QA Engineer does in an unusual case

Some rules are written as “if”: they fire not in every conversation but in the one that matches. That answers the unspoken “what if my case is different” and removes the need to reshape the task to fit the role.

“If the feature is described vaguely, first return the questions without which testing is meaningless.”

Three questions QA Engineer offers first

A click opens the chat with the message already typed. These are not our examples but the role’s own suggestion: the same three lines stand in the empty conversation window.

Write test cases for a form

Where will this break?

A regression checklist before release

Questions about this assistant

Quick answers

Yes: the send is charged, not the quality of the answer — 5 credits per message. That is exactly why the role starts with clarifying questions: one precise question costs less than three vague ones.

ChatGPT GPT-5.5. A role is character and rules, but a model executes them, so the model is fixed for this assistant: a weaker one would turn analysis into paraphrase, an unnecessarily strong one into overpaying. You can switch models inside the chat.

Both roles sit on the “Code and technology” shelf but answer by different rules: the rules of QA Engineer are quoted in full on this page, and those of “Prompt Engineer” on its own. The full shelf is listed at the end of this page.

Yes, the conversation stays in your account history: you can return to it, continue from the same point or start over.

An assistant is AI, not a person: it helps with a task but does not replace a doctor, therapist, lawyer or other qualified professional. Verify important decisions and facts against a primary source.

How QA Engineer differs from a plain chat

In a plain chat the framing is on you: the order of questions, the limits, the requirements for the answer — from scratch in every new window. Here it is already written: 7 rules, 2 prohibitions of them, all quoted on this page. You never have to restate them.

✓ Rules: 7, prohibitions: 2
✓ Catalogue shelf: Code and technology
✓ A program answers, not a person: it replaces no doctor, lawyer or therapist