50 QA prompts, each in two formats, XML-tagged for Claude, Markdown-structured for GPT, covering test design, automation, security, accessibility, CI/CD, and AI agent workflows.
Given one requirement, the model returns a structured suite of 8-10 test cases, ID, title, preconditions, steps, expected result, priority, and the design technique that generated each case, covering happy path, negative scenarios, boundary values, equivalence partitions, and edge cases.
Reads a user story and returns Cucumber-compatible Gherkin scenarios covering happy path, error scenarios, and at least two edge cases. Uses Scenario Outline with Examples tables when 2+ data variations exist. Forbids UI implementation detail in Then steps.
Returns a Session-Based Test Management charter for a feature area, including a focused mission statement, in-scope areas, applicable testing heuristics by acronym (FEW HICCUPPS, SFDIPOT, RIMGEA), timebox, environment, and a structured notes template.
Reads a data schema (JSON Schema, TypeScript type, or table description) and returns four explicit fixture groups: happy path, boundary, invalid, and edge cases. Edge group must include Unicode, RTL, emoji, and injection payloads. Output is a ready-to-import JSON file.
Reads field constraints (min/max/length/format) and returns boundary and equivalence-partition cases for each, values just inside, on, and just outside the boundary, with technique labels (BVA or EP) per case.
Give the model a page description plus a list of UI elements and it returns a complete Page Object Model in TypeScript using Playwright's auto-waiting locators (getByRole / getByTestId), typed action and assertion methods, and a page-level fixture.
Returns a Cypress Page Object class using cy.get() with data-cy preference, action methods that chain via `return this`, plus a commands.js file for cross-page utilities and a sample spec consuming the POM.
Reads manual test steps (Action / Expected Result) and produces a Playwright spec with locator suggestions, action method calls (assuming a POM exists), assertions matching expected results, and explicit `// MANUAL:` comments where automation can't replicate human judgment.
Reads an OpenAPI 3.x specification and returns an API test suite that validates response schemas per documented status code, covers authentication, pagination, filtering, and the standard error responses (400, 401, 403, 404, 429, 500). Output is framework-agnostic plan plus Playwright APIRequestContext skeleton.
Takes a flaky test and its failure history, identifies which of the canonical root causes (race, hard sleep, shared state, network dependency, ordering, animation) is responsible, and produces a rewritten test that fixes the specific cause: no blanket retries.
Returns a full Test Strategy document with 8 mandatory sections, scope, test levels, test types, entry/exit criteria, risk matrix, environment, resources, deprioritized test types, tailored to a product description and stated constraints.
Reads a Product Requirements Document and returns a release-specific Test Plan with scope, milestones, RACI per major activity, deliverables, entry/exit criteria, risk assessment, and a defect management workflow. Uses date placeholders the user fills rather than fabricating.
Reads a feature list and outputs a prioritization matrix with four weighted dimensions (business impact, technical complexity, usage frequency, defect history) scored 1-5, total weighted score, recommended test execution order, and tie-break rule.
Reads a microservices topology (services + integration points) and returns a recommended pyramid distribution with concrete percentages per layer, layer-by-layer justification, explicit contract-test placement at every service boundary, and a strategy for async events.
Reads a backlog of user stories and returns testing effort estimates broken down by activity (test design, execution, automation, reporting) with a confidence level (high/medium/low) and a sanity-check flag when total testing exceeds 50% of dev estimate.
Takes a free-form issue description (Slack message, email, support ticket) and returns a structured bug report following the AQAPro Bug Report Template: clear `[Component] Verb-noun` title, environment, separate severity and priority, numbered atomic repro steps, expected vs actual, and suggested investigation areas.
Given a defect description, returns a literal 5 Whys chain, a fishbone diagram (text representation) categorizing contributing factors into People / Process / Technology / Environment, and a list of preventive measures with named owners: never generic recommendations.
Reads a bug description and assigns SEVERITY (impact on system, 1-4) and PRIORITY (urgency to fix, 1-4) on independent scales, each with a written justification, plus a recommended SLA target. Refuses to collapse the two dimensions into one score.
Reads application logs, HAR files, or browser console excerpts and reconstructs a step-by-step reproduction recipe with timestamps, the failing request, suspected preconditions, and a confidence flag per inferred step.
Given a new bug description and N existing bug summaries, returns a ranked list of duplicate candidates with similarity scores (0-100) based on ROOT-CAUSE likelihood rather than surface text: with one-line evidence per candidate.
Returns a load test strategy covering 5 scenario types (baseline / load / stress / spike / soak) with thresholds for response time, throughput, and error rate, environment requirements, monitoring checkpoints, and pass/fail criteria, and explicit environment-parity statement.
Reads an endpoint description and returns a ready-to-run k6 script with `options.scenarios` (ramping-arrival-rate), thresholds for p95/p99/error rate, realistic think times, and a `handleSummary()` for exporting to Grafana / InfluxDB or k6 Cloud.
Returns a JMeter test plan as a valid .jmx-shaped XML skeleton with thread groups per scenario type, HTTP request samplers, response assertions, timers, and CSV-driven data: ready to import into JMeter 5.x for refinement and distributed runs.
Reads a load test result summary (latency percentiles, throughput, error rate, system metrics) and returns a ranked list of suspected bottleneck layers, network, application, database, dependent service, or infrastructure, each with evidence cited from the metrics and a recommended next investigation step.
Reads a critical user journey and returns a Playwright-based synthetic monitoring script with business-step checkpoints, failure-screenshot capture, an alerting threshold tied to a stated SLO/SLI, and a recommended run frequency.
Returns an OWASP ASVS-aligned security testing checklist covering authentication, session management, authorization, input validation, output encoding, cryptography, API security, file upload, and HTTP security headers: each item with a test method (manual / DAST / SAST) and ASVS chapter citation.
For each OWASP Top 10 (2025) category (A01-A10), returns 3-5 concrete test cases with payloads, recommended tools, expected secure behavior, and remediation guidance tailored to the target application.
Returns rate-limiting test cases verifying X-RateLimit-* headers, behavior at the limit boundary, burst handling, reset window semantics, per-authentication-level limits, and concurrent request behavior across multiple workers.
Returns a structured suite of authentication and authorization bypass test cases, IDOR, JWT algorithm confusion, session fixation, MFA bypass, brute-force resistance, broken object-level authz, with payloads, CWE numbers, and the detection signal that confirms vulnerability vs secure behavior.
Reads an SBOM (CycloneDX or SPDX) and returns test cases for known-vulnerability presence (CVE lookup against KEV catalog), license compliance (allow/deny list), transitive dependency drift, and prioritized update list based on exploitability and reach.
Returns a comprehensive WCAG 2.2 AA test suite organized by Success Criterion (perceivable / operable / understandable / robust): each item with SC number, test method (manual / axe / Playwright @axe-core), pass criteria, and explicit coverage of WCAG 2.2's new criteria.
Reads a component description and returns screen reader test scenarios with expected announcement text per reader (NVDA, JAWS, VoiceOver), recommended browser pairing per reader, navigation pattern, and clear pass/fail criteria.
Returns numbered keyboard navigation test cases covering Tab order, focus visibility, focus trap in modals/menus, Escape key behavior, skip links, focus return after modal close, and roving tabindex for composite widgets: with expected behavior per case.
Reads a list of foreground/background color pairs and returns contrast ratios, WCAG 1.4.3 (text) and 1.4.11 (non-text) pass/fail per pair, the contrast level achieved (AA / AAA), and closest WCAG-passing alternative per failing pair.
Reads a component's ARIA implementation (role + states + keyboard handling) and validates against the matching WAI-ARIA Authoring Practices Guide pattern: flagging missing required roles, missing required states/properties, incorrect keyboard interactions, and prohibited attribute use.
Returns a complete `.github/workflows/qa.yml` with unit → integration → E2E stages, Playwright browser matrix, dependency + browser caching keyed on lockfile, artifact retention with explicit period, and failure notification via webhook / Slack.
Returns a `.gitlab-ci.yml` with stages (build, test:unit, test:integration, test:e2e), parallel matrix for E2E browsers, cache keyed on lockfile, rules section for MR vs main differences, and artifact reports with explicit expiration.
Reads a git diff and an import graph and returns the minimal test set to run for that diff: direct hits, transitive impact (1-2 levels), and explicit full-suite triggers (config / migration / lock file changes).
Reads a test file and annotates each test with Allure metadata (epic, feature, story, severity, link to spec, TMS link) based on file path conventions, commit message JIRA prefixes, and test naming: emits a diff, not a rewrite.
Returns a parallel-execution config tailored to your framework (Playwright or Jest), CI runner count, average test duration, and flakiness rate, including shard count, worker count per shard, test ordering strategy, and a reasoning paragraph.
Reads a test file and returns a categorized list of anti-patterns: hard sleeps, shared mutable state, weak assertions (`toBeTruthy` instead of `toEqual`), missing teardown, mixed setup/assertion concerns, each with line numbers, severity, and a suggested fix.
Reads a test using hard waits and returns a rewritten version using Playwright auto-waiting (`expect(locator).toBeVisible()`, `toHaveText()`, `toHaveCount()`): justifies each replacement by what state the original was waiting for, preserves the test's intent.
Returns a per-test-file quality checklist with 20-30 items grouped by category (naming / structure / assertions / isolation / performance / maintainability): each marked PASS/FAIL with one-line evidence from the code.
Scans a set of test files and identifies duplicated setup, fixture state, and assertion patterns: proposes refactors using Playwright fixtures, factory functions, or shared helper modules with concrete code diffs. Warns against premature abstraction (single-use helpers).
Reviews a Page Object Model class and returns specific refactoring suggestions: locator priority (role > label > testid > CSS), action vs assertion separation, action granularity (one method per user intent), and constructor cleanliness: with diff-style proposed changes.
Reads a description of your QA project (framework, language, conventions, CI setup) and returns a ready-to-commit `CLAUDE.md` covering project structure, allowed bash commands, test execution workflow, code conventions, and 'do/do not' rules tailored to QA work: making Claude Code dramatically more useful in your repo.
Returns a `.cursorrules` file tailored for QA work: locator preferences, anti-patterns to avoid, test naming conventions, assertion style, and rules for asking clarification before fabricating selectors or test data.
Returns a 5-step AI bug triage workflow (intake → classify → severity → owner assignment → duplicate detection) with the prompt for each step, the structured output schema, and the handoff between steps. Final step emits a structured triage decision.
Reads a Figma design's structured description (frames, interactions, prototype links, states) and returns user-flow test cases plus a Playwright spec scaffold with `data-testid` suggestions matched to design layer names. Flags missing interaction states.
Returns a prompt + decision framework for an AI PR-review bot tuned for QA: checks for test anti-patterns, missing tests for new branches/conditions, naming convention violations, and assertion quality, with explicit rules for when to comment vs skip and how to phrase comments without false confidence.
AI prompts for QA are reusable, structured instructions that get an LLM (Claude, GPT, or Gemini) to produce QA artifacts, test cases, Playwright page objects, bug reports, OWASP scenarios, and CI configs, consistently. A good prompt encodes the format, constraints, and edge cases so output is reliable.
Yes. All 50 prompts on AQAPro are free to copy and use. Every prompt ships in two formats, Claude (XML-tagged) and GPT (Markdown-structured), and a couple also carry a Gemini variant. No signup and nothing is uploaded.
Use the variant that matches your assistant. Claude variants use XML tags, because Claude follows tagged structure well; GPT variants use Markdown headings. Both end with a short Approach step that says what to settle before writing. There is no "think step by step" scaffolding: current Claude, GPT and Gemini models reason on their own, and how hard they think is a model or effort setting, not something a prompt needs to ask for. The task and constraints are identical: only the structure differs. The prompts aren't tied to a model version, so they work with whichever model your assistant runs today. Where a Gemini variant exists, a third tab appears on that prompt's page.
No. Prompts accelerate drafting test cases, scripts, and reports, but a QA engineer reviews coverage, validates edge cases, and owns risk decisions. Treat prompt output as a fast first draft.