Blog

Set up Playwright MCP with Claude Code, run it in CI, and store the results across runs. A complete guide to the browser-automation agentic loop.

Is Playwright free? A 2026 Playwright pricing breakdown of Microsoft Playwright Testing, Currents, and test-result hosting, compared by billing model.

A fair comparison of 8 test automation reporting tools in 2026: Gaffer, Allure, ReportPortal, BrowserStack, ExtentReports, Testmo, Katalon TestOps, and TestRail.

Generate Playwright HTML, JSON, JUnit, and Blob reports, run them in CI, and host them so your team can open every run without the local machine.

What a test report should contain, the types of test reports, and how small dev teams automate test reporting in CI instead of writing reports by hand.

How to use Playwright's planner, generator, and healer agents (v1.56+), run them in CI, and track whether agent-written tests actually stay stable over time.

Generate Cypress test reports with mochawesome, Allure, JUnit, or CTRF reporters, wire them into CI, and host them so your team can see results.

When 200 tests fail in CI, there are rarely 200 problems. A guide to test failure types, what causes each, and how to cluster by root cause.

I tried to fix flaky Playwright tests by using Gaffer the way a coding agent would: find flakies, narrow with `affected-tests`, fix, verify. The loop broke at step two. Here's what was wrong and what shipped.

Wire Playwright MCP into Claude Code, then close the loop with Gaffer's MCP so the agent reads CI results, spots flaky failures, and iterates.

GitHub Agentic Workflows let AI agents run as CI. Here's how we connected Gaffer's MCP server for automated weekly test health reviews.

Per-build notifications tell you a test failed. Health score alerts tell you your suite is degrading. Here's how they work.

Most teams retry and move on. A systematic approach to flaky test management: flip rate detection, CI cost prioritization, and a fix-quarantine-delete framework.

Your observability stack monitors production but not tests. Gaffer's OpenTelemetry export puts test metrics into Datadog, Grafana, and any OTLP endpoint.

AI coding tools know your code but not your tests. Test intelligence via MCP bridges the gap with flaky detection, failure clustering, and coverage data.

Claude Code, Cursor, and Windsurf can query your test history, find flaky tests, and diagnose failures — if you give them structured data. Here's how.

Agents write code and run tests. But when tests fail, they're flying blind. Here's how test intelligence closes the loop.

Flaky tests are an invisible tax on engineering velocity. Industry data from Atlassian, Google, and Microsoft shows 13-16% of test failures are flaky. Here's how to calculate the real cost.

Dogfooding Gaffer's MCP coverage tools on our own codebase.

JUnit XML is the de facto test report format used by Playwright, Jest, Vitest, pytest, Mocha, and Go. Schema, working examples, and CI integration patterns.

Test artifacts are the files generated when tests run: reports, logs, screenshots, coverage data. Here's what they are and how to manage them at scale.

Using Gaffer's own analytics to find and fix slow E2E tests. Here's what I found and how I fixed it.