Replay QA

Replay QA Overview

Replay QA is an autonomous quality assurance agent. Give it a URL or a GitHub repo. It explores your app like a human, records every session, and delivers root-cause analysis with suggested fixes.

How it works

Create a project

Go to qa.replay.io/new and provide either a live project URL or a GitHub repository URL. Replay QA uses this as the starting point for testing.

Explore

Replay QA navigates your application by clicking through pages, opening menus, and filling in forms. It maps the journeys users can take through the app. For applications with accounts and checkout flows, it attempts signup, login, cart interactions, and payment.

Test

Each journey runs in a real browser. Replay QA writes Playwright tests for each flow as it goes. No test suite or configuration is required upfront.

Record

Every test session is recorded with the Replay time-travel debugger. You can replay the exact moment a bug occurred, inspect network requests, read console output, and trace execution.

Project settings

Polish passes

Replay QA can run additional polish passes alongside functional testing. Each pass checks a specific category and can be toggled on or off for each project.

PassDescription
Network PerformanceChecks for slow API responses, large payloads, and excessive requests.
React RenderingIdentifies unnecessary re-renders and component performance issues.
Layout ShiftDetects elements that move after page load, causing cumulative layout shifts.
AccessibilityAudits for missing ARIA labels, contrast issues, and keyboard navigation gaps.
GlitchesLooks for visual artifacts, flickering, and broken animations.
User ExperienceEvaluates form validation feedback, loading states, and error handling.
UI DetailsChecks alignment, spacing, and visual consistency across pages.

Design document

Replay QA can test against a design document that describes the intended behavior of your application. Provide details about what your app does, its main features, key user flows, and any specific expectations QA should test against. The more detail you provide, the better QA can target its testing.

Logins

You can add or change multiple sets of login credentials. They are saved as the project login variables and used to test signed-in areas.

Issue tracking

File bugs directly to your issue tracker. Each issue includes the full report: root cause, steps to reproduce, screenshots, and the Replay recording.

Supports GitHub Issues, Jira, and Linear. Configure the integration in your project settings to pick where QA should file the bugs it finds.

Report

Every bug report includes:

FieldDescription
Root causeWhy the bug happened, grounded in recorded execution data
Suggested fixCode-level recommendation for fixing the bug
Full recordingLink to Replay DevTools session with network, console, and DOM state
Confidence scoreHow reliably the bug was reproduced (high / medium / low)

Each bug report also includes an agent-ready description of the problem and fix: a structured report with title, description, steps to reproduce, expected behavior, root cause, causal chain, and recording links that can be fed directly into an AI coding agent.

Public apps

If an application is posted to a public link (Show HN, Reddit, X), it can be tested for free. Create a project with the URL and any required logins, open the project settings, and set the public share link to the post. The project becomes free immediately and any existing credits are refunded.

Shared projects are public and searchable, allowing others to run QA passes on them as well.

Get started

Test a URL

Paste a link and get a bug report in minutes.

Connect a GitHub Repo

Set up continuous QA that runs on every change.

Previous
FAQ