Agent Claude Code
Use to audit a web UI for accessibility against WCAG 2.2 AA. Runs axe-core on key pages and components, then performs the manual keyboard, focus, and screen-reader review that automation can't — driving the live app via the Playwright MCP when available — and produces a WCAG-referenced report where every finding names…
Agent Claude Code
Use to add or extend API contract tests for a service. It detects the stack and interface (OpenAPI/GraphQL/Pact), recommends consumer-driven vs spec-first, scaffolds the tests, and wires can-i-deploy / breaking-change gates into CI.
Agent Claude Code
Use to triage a suspected flaky test end to end. It reproduces the non-determinism by rerunning the test many times (and varying order, workers, timezone, and seed), classifies the root cause against the flake taxonomy, then either proposes a minimal root-cause fix or quarantines the test with a required owner…
Agent Claude Code
Use to audit and improve how effective a project's Jest unit tests are. It sets up/runs coverage and Stryker mutation testing, identifies survived mutants and coverage gaps, then strengthens tests to kill the mutants and wires CI gates. Give it a module or the whole src to focus on.
Agent Claude Code
Use to design and build LLM/RAG/agent evaluation suites in DeepEval that gate a release on output quality. It elicits or derives the golden dataset and the failure mode to guard against, picks the metrics that match it (faithfulness/answer-relevancy for the generator, contextual precision/recall for the retriever…
Agent Claude Code
Use to design and run performance tests that prove a system meets its SLOs under realistic load. It elicits or derives SLOs and a workload model, writes k6 load/stress/soak scripts and Lighthouse budgets with thresholds-as-gates, runs them against a production-like target, and interprets the results against the SLOs …
Agent Claude Code
Use to convert existing imperative Playwright tests into BDD with the playwright-bdd runner. It recovers each test's business intent, writes declarative Gherkin .feature files (business language, no clicks/selectors), extracts page objects, wires step definitions to Playwright fixtures, and verifies behavior parity.…
Agent Claude Code
Use to author or extend Playwright end-to-end tests for a user journey. Give it the flow to cover; it produces Page Object Model specs with stable locators, web-first assertions, fixture-based isolation, and storage-state auth — and can drive a live browser via the Playwright MCP server to inspect the real UI before…
Agent Claude Code
Use to create a tailored QA strategy for a team or project. It runs a structured intake (tech stack, team size, release cadence, current maturity, risk/compliance), then produces a risk-based strategy with an automation plan, quality gates, tooling, and a phased roadmap. Enforces guardrails against invented inputs…
Agent Claude Code
Use to analyze a full-stack application (frontend, backend, middleware) and produce a complete test pyramid strategy — what to test in the FE, what in the BE, what at the middleware/seams, and at which level. It inspects the codebase, maps the layers, and writes a per-layer test plan with tooling and CI wiring.
Agent Claude Code
Use to audit a UI/E2E test suite for overuse and recommend which tests should move to the API or unit layer. Scans Playwright, WebdriverIO, and Selenium/WebDriver tests in any language (TS/JS, Python, Java, C#, Ruby), finds data-driven repetition and logic tested through the browser, and writes a per-test migration…
Agent Claude Code
Use to set up or repair Playwright visual regression tests for a UI. Give it the views or components to cover; it makes screenshots deterministic (disable animations, freeze the clock, seed data, pin fonts/viewport/device-scale), adds a stabilization fixture, chooses masking and thresholds, establishes container-based…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: