Skip to content

Prompts

Jump to bottom
github-actions[bot] edited this page May 18, 2026 · 1 revision

System Prompts

6 battle-tested system prompts for AI-powered QA. Browse them in the live marketplace and click "Copy System Prompt" to grab any one.

PRD to Test Matrix

Tagline: Convert requirements into exhaustive test plans

Analyzes a PRD or user story and generates a comprehensive Test Plan with risk assessment, scope definition, a full test matrix (functional, negative, edge cases), NFR scenarios, and a Playwright automation strategy.

You are an elite Lead QA Architect with over 15 years of experience in enterprise software testing.
You possess a hacker's mindset and a deep understanding of complex system architectures...
[full text in UI]

Output format: Markdown with Feature Overview, Scope, Test Matrix tables, NFR section, Automation Strategy

The Hacker QA

Tagline: Generate malicious & boundary JSON test data

Produces a raw JSON array of attack payloads: SQL injection, XSS, nulls, RTL text, emoji, boundary numbers, oversized strings, and Unicode control characters.

You are an elite Senior QA Data Engineer.
Generate JSON test data designed to break systems — SQLi, XSS, nulls, RTL text, emoji, boundary numbers.
Return ONLY a raw JSON array. No markdown fences. No commentary.

Output format: Raw JSON array with scenario_description field on each object

PRD Analyzer (Quick)

Tagline: Fast PRD → test matrix with sample API payloads

Faster than the full PRD to Test Matrix prompt — produces a Markdown table matrix plus sample JSON payloads for each edge case.

You are a meticulous QA Architect. I will provide you with a PRD or a feature description.
1. Analyze: Identify implicit requirements, contradictions, missing error-handling...

The BDD Master

Tagline: PRD → Gherkin/Cucumber scenarios (Given/When/Then)

Produces a complete .feature file suite with @smoke/@regression tags, Background, happy path, negative Scenario Outlines with Examples tables, and @nfr scenarios.

Coverage per feature (minimum):

  • 1 happy path with concrete data
  • 3 negative scenarios (invalid input, unauthorized, downstream failure)
  • 2 edge cases (boundary values, concurrent actions)
  • 1 @nfr scenario

Strict SDET PR Reviewer

Tagline: Ruthless Playwright/Cypress PR review

Reviews PR diffs for flaky locators, missing assertions, hardcoded waits, secrets in tests, and no test isolation. Outputs blockers, warnings, praise, and a 1–10 flakiness score.

Blockers checked:

  • CSS/XPath with dynamic IDs or nth-child
  • Actions without expect() on outcome
  • waitForTimeout, cy.wait(5000)
  • API keys/passwords in source
  • Shared mutable state between tests

API Contract Enforcer

Tagline: OpenAPI/Swagger → comprehensive API test matrices

Given an OpenAPI spec, generates a parameter permutation matrix, status code coverage plan, schema validation cases, copy-paste HTTP sample requests, and automation hints for Playwright APIRequestContext.

Covers: required vs optional omission, type violations, boundary values, all auth variations, security cases (SQLi in query params, oversized payloads)

Clone this wiki locally