qa-suite

Turns user stories, PRD/SRS documents or tickets into atomic, testable requirements with REQ IDs, ISO 29148 quality review, contradiction and gap detection, risk scores and clarification questions. Use when requirements need review before test design. Triggers include requirements analysis, user story review, acceptance criteria, missing requirements; Turkish "gereksinim analizi", "gereksinimleri incele", "kabul kriteri yaz".

Turns QA Suite test cases into Playwright TypeScript automation that keeps each TC ID as a tag and each step as test.step, scaffolds config, page objects, auth and CI, and feeds results back by TC ID. Use when designed tests should become UI automation. Triggers include Playwright, test automation, e2e UI tests, automate manual tests; Turkish "otomasyon yaz", "Playwright testleri", "test case'leri otomatize et".

Designs risk-based manual test cases from requirements with ISTQB techniques; boundaries, partitions, decision tables, state transitions and pairwise sets are computed by scripts, with stable TC IDs. Use when someone needs test cases or asks what to test. Triggers include write test cases, test scenarios, boundary value analysis, decision table, negative tests; Turkish "test case yaz", "test senaryosu çıkar", "sınır değer analizi".

Exports QA Suite test-cases.json into validated import files for Xray, Zephyr Scale, TestRail, Azure DevOps and Qase, or an Excel workbook, CSV or Markdown document, keeping Jira requirement links and Turkish characters intact. Use when test cases must move into a test management tool. Triggers include Xray, Zephyr, TestRail, Jira import file, export to Excel; Turkish "Xray'e aktar", "Jira'ya import", "Excel'e dök".

Writes a risk-based test plan for a release or feature (ISO/IEC/IEEE 29119-3) with scope, risk register, approach, measurable entry/exit criteria, environments, schedule and an effort estimate computed from QA Suite artifacts. Use when testing must be planned or estimated. Triggers include test plan, test strategy, entry and exit criteria, test estimation; Turkish "test planı hazırla", "test stratejisi", "test eforu tahmini".

Prepares test data by deriving each test case's needs, generating deterministic synthetic CSV/JSON with referential integrity, edge values and valid Turkish IDs (TCKN, VKN, TR IBAN, +90 phones), and masking production extracts under KVKK/GDPR. Use when tests need seed, fixture or bulk data. Triggers include test data, synthetic data, fake customers, data masking; Turkish "test verisi hazırla", "sahte veri üret", "veri maskeleme".

Runs the whole QA Suite chain from requirements through analysis, risk-based test design and gap checks to tool export and automation, with one set of stable IDs. Use only when someone wants a complete end-to-end QA package, not a single stage or test type. Triggers include end-to-end test analysis, full QA package, whole QA workflow; Turkish "baştan sona test analizi", "QA paketi hazırla", "test sürecini yürüt".

Writes defect reports (severity vs priority, evidence) and computes test status and completion reports with pass rate, per-requirement verdicts, open defects, residual risks and exit-criteria verdicts. Use when results must become a bug report or a release decision. Triggers include bug report, defect report, test summary report, release readiness, QA sign-off; Turkish "hata raporu yaz", "test özet raporu", "yayına hazır mıyız".

Reviews an existing test suite written by people or AI. Imports Excel (.xlsx) or tool CSV exports, flags vague or missing expected results, missing data, multi-action steps and unlinked tests, and applies a rubric for oracles, depth and risk coverage. Use when test cases need a quality audit. Triggers include review test cases, test case quality, Excel test import; Turkish "test case'leri incele", "bu testler yeterli mi".

Plans and reports session-based exploratory testing (SBTM). Writes risk-ranked charters with heuristics and time boxes, guides tagged-note sessions, also by an AI agent in a browser, and turns session notes into metrics, defect drafts and draft regression tests. Use when an app should be explored for unknown bugs. Triggers include exploratory testing, charters, bug hunt, test tours; Turkish "keşif testi", "keşifsel test", "uygulamayı gez ve hata bul".

Tests LLM and generative-AI features (chatbots, RAG, agents) with a TR/EN eval dataset of functional and OWASP LLM Top 10 cases (prompt injection, excessive agency), scoring repeated runs for pass rate, flakiness, latency and cost with a release gate. Use when an AI feature must be evaluated. Triggers include LLM evals, chatbot testing, RAG evaluation, hallucination, jailbreak; Turkish "yapay zekâ testi", "chatbot testi", "prompt injection testi".

Tests REST APIs from their OpenAPI 3.x contract. Generates traceable test cases and an executable API suite with the same TC IDs (happy path, schema, 401, required fields, boundaries, enums, 404), then guides business-rule and idempotency tests. Use when endpoints or a Swagger document need testing. Triggers include API testing, REST endpoints, contract tests, IDOR/BOLA authorization; Turkish "API testi", "Swagger'dan test çıkar", "endpoint'leri test et".

Tests data migrations, ETL pipelines and system replacements against the mapping specification. Profiles source data (duplicates, cp1254 vs UTF-8, decimal comma), reconciles source and target by script (counts, keys, field mismatches, control totals) and rehearses cutover and rollback. Use when data moves between systems. Triggers include data migration, ETL testing, reconciliation, cutover; Turkish "veri göçü testi", "migrasyon testi", "mutabakat".

Tests native and hybrid iOS/Android apps. Drafts test cases for lifecycle, upgrades, offline, permissions, push, deep links, payments, OWASP MASVS and mobile accessibility (TalkBack/VoiceOver), builds a device matrix and maps Maestro/Appium/Espresso/XCUITest results by TC ID. Use for mobile apps and store releases. Triggers include mobile testing, device matrix, App Store or Google Play release; Turkish "mobil uygulama testi", "iOS ve Android testi", "cihaz matrisi".

Designs non-functional tests traced to requirements. Generates k6 performance scripts with thresholds from the requirement (p95, error rate), WCAG 2.2 AA accessibility checks and OWASP ASVS 5.0 security tests selected by feature. Use when performance, accessibility or security needs testing. Triggers include load testing, k6, WCAG, a11y, security testing; Turkish "performans testi", "yük testi", "erişilebilirlik testi", "güvenlik testi".

Builds a requirements traceability matrix (RTM) from QA artifacts, measures coverage, reports risk-ordered gaps (uncovered, no negatives, orphans) and change impact, and does risk-based regression selection within a time budget. Use when someone asks about coverage or what to re-test. Triggers include RTM, test coverage, traceability, change impact, regression selection; Turkish "izlenebilirlik matrisi", "test kapsamı", "regresyon seçimi".

Writes BDD Gherkin scenarios (.feature files, TR/EN) from QA Suite test cases or story criteria, keeping each TC ID as a tag, in declarative business language, runnable through playwright-bdd with shared page objects. Use when executable specifications or living documentation are wanted. Triggers include Gherkin, BDD, Cucumber, Given/When/Then, feature files; Turkish "Gherkin yaz", "BDD senaryosu", "Diyelim ki / Eğer ki / O zaman".