Exploratory QA Specialist

What it does

Exploratory QA Specialist is a Cursor agent playbook named qa-explorer. AgenticKit installs it so a session can adopt this role instead of acting as a generic coding assistant.

The qa-explorer agent uses the product like a stranger. test-automator writes suites first; this role finds what suites miss: confusing copy, dead buttons, checkout dead-ends. Output is a ranked bug list, then optional tests.

Clicks the app like a new user and files what breaks. Runs a human pass on a flow, lists bugs and confusions with severity and repro. Writes Playwright only after findings.

Why it exists

A pile of agent files is not a team. qa-explorer exists so one job, exploratory qa specialist, has a written brief, required skills, and a stop condition.

It ships in the Engineering Kit. Skills it must read: Exploratory QA Heuristic Pass; Playwright End-to-End Browser Testing. Commands that adopt this role: /qa-pass: Exploratory QA Pass; /e2e-suite: Playwright End-to-End Test Suite Generator; /onboarding: First-Run Activation & Empty States.

When to use it

• Before a launch when you need fresh eyes
• Checkout that 'works' but people bounce
• A PR that needs a click-through, not more unit tests

When not to use it

• Do not ask qa-explorer to do a different role's job. If you need a launch post, switch agents.
• Do not keep the same chat after this agent has finished its artifact. Start a reviewer in a new thread.
• Do not invoke every agent in the kit for a small change.

Example workflow

1. Install the Engineering Kit so .cursor/agents/qa-explorer is on disk.
2. Run /qa-pass: Exploratory QA Pass, or start a chat and tell Cursor to adopt the qa-explorer role.
3. The agent should read: Exploratory QA Heuristic Pass; Playwright End-to-End Browser Testing.
4. It produces the artifact for this role only, then stops.
5. A different agent or you review. Same-chat self-review is not a review.

Example usage

Example: "You are qa-explorer. Before a launch when you need fresh eyes. Read exploratory-qa before you edit."

Or let the pipeline invoke it: /qa-pass: Exploratory QA Pass.

Capabilities you should actually see: Heuristic pass (clarity, errors, mobile, empty states)

Example output

• qa-explorer should leave files or a written verdict, not a vibe check. Heuristic pass (clarity, errors, mobile, empty states) Severity-ranked findings with repro
• Engineering agents should touch the slice they were given (schema, route, test, or review note) and nothing else.

Best practices

• Heuristic pass (clarity, errors, mobile, empty states)
• Severity-ranked findings with repro
• Confusion notes that are not bugs but lose users
• Optional Playwright after the human pass
• One role per chat unless a command is explicitly orchestrating a sequence.

Common mistakes

• Using qa-explorer as a synonym for 'the Cursor agent'. It is a brief, not the product.
• Skipping the required skills and hoping the role name is enough.
• Letting the writer approve its own PR.
• Invoking this agent and three unrelated ones in the same prompt.