Revyl

Product Introduction

1. Definition: Revyl is a cloud-based mobile testing and observability platform designed for engineering, QA, and product teams. It provides real-time access to live iOS and Android cloud devices and generates a comprehensive runtime record of app behavior.
2. Core Value Proposition: Revyl exists to eliminate the "verification lag" in fast-paced mobile development by providing a single source of truth built from runtime evidence. It delivers mobile test automation, performance monitoring, and runtime context to help teams ship with confidence.

Main Features

1. Live Device CLI & Infrastructure: Revyl provides command-line interface (CLI) and Model Context Protocol (MCP) access to provision live cloud devices in seconds. It establishes a low-latency (<50ms) WebRTC stream at 60fps, enabling hot reload, tap, type, swipe, and screenshot interactions. This feature allows AI agents and developers to interact with real device environments, installing builds and verifying changes in a live loop.
2. Agentic Mobile Testing (Natural Language Automation): Teams can define end-to-end mobile workflows using natural language. The platform's AI agents execute these workflows on real iOS and Android builds, performing actions and assertions. This creates automated mobile test suites that catch regressions with replayable evidence, moving beyond traditional script-based UI testing.
3. Replayable Evidence & Runtime Record: Every test execution and device session generates a complete evidence package. This includes session video, high-resolution screenshots, device logs, network waterfalls, CPU/memory/FPS performance data, step-level execution traces, and AI reasoning. This evidence is shareable via a URL, providing immutable proof for debugging, design review, and product validation.
4. Atlas Runtime Mapping: Atlas is a core innovation that automatically maps an application's structure from observed tests and sessions. It creates a living runtime map of screens, flows, UI states, and navigation paths, serving as an up-to-date source of truth for the app's actual behavior in production-like environments.

Problems Solved

1. Pain Point: Verification Lag & Runtime Context Gap. Mobile teams using CI/CD often lack visibility into how builds behave on real devices. Traditional testing may verify element presence but misses performance issues, network behavior, and stateful interactions. This leads to "What broke?" becoming a costly, time-consuming question, causing flaky tests, reruns, and late-stage bugs that drag down release cycles.
2. Target Audience: The primary users are Mobile Engineers (iOS/Android), QA Automation Engineers, Designers, and Product Managers at companies shipping apps built with React Native, Flutter, Expo, or native Swift/Kotlin. It is also built to provide device access and evidence for AI development agents.
3. Use Cases:
   • Continuous Integration Verification: Running a regression suite across multiple cloud devices on every build via GitHub Actions to catch platform-specific bugs.
   • Performance Debugging: Using CPU, memory, and FPS traces to pinpoint performance regressions in a specific app flow.
   • Design & Product QA: Comparing a build against design specs using replayable evidence and the live app view, not just static mockups.
   • AI Agent Development: Providing agents with a controlled, observable mobile environment to develop and test autonomous workflows.

Unique Advantages

1. Differentiation: Unlike traditional cloud device farms (e.g., BrowserStack, Sauce Labs) that focus on manual or scripted testing, Revyl emphasizes creating a unified runtime record and automated observability. It combines testing, evidence collection, and app mapping into a single platform, contrasting with fragmented toolchains (one tool for testing, another for monitoring, another for device access).
2. Key Innovation: The core innovation is the Atlas Runtime Map and the concept of a "runtime record." Revyl doesn't just run tests; it observes all interactions to build an always-current model of the app. This turns every test and session into a data point that enriches the team's shared understanding of the application's live behavior, fundamentally shifting from "did the test pass?" to "what did the app actually do?".

Frequently Asked Questions (FAQ)

1. How does Revyl mobile testing work with CI/CD pipelines like GitHub Actions? Revyl integrates directly into CI/CD workflows via its CLI or the official revyl/run-tests GitHub Action. You can trigger parallel test runs across multiple iOS and Android cloud devices on every commit or build, receiving detailed pass/fail reports and evidence links directly in your pipeline logs.
2. What mobile frameworks and platforms does Revyl support? Revyl supports React Native, Expo (managed and bare), Flutter, and native iOS (Swift) and Android (Kotlin) applications. You provide a build artifact (.APP or .APK), and Revyl runs it on its cloud infrastructure, which includes iOS cloud simulators and Android cloud emulators.
3. How is Revyl different from traditional mobile test automation tools like Appium or XCUITest? Traditional tools focus on script execution for UI validation. Revyl incorporates the test execution into a broader observability platform. It automatically captures rich performance data, network logs, and state diffs alongside test results, providing full context. Furthermore, Revyl enables natural language test definitions and maintains a living app map (Atlas) from test executions.
4. Can I use Revyl to test and control devices during my local development cycle? Yes. Revyl's Live Device CLI is designed for the local dev loop. Developers can spin up a cloud device session from their terminal, install a build via hot reload, and interact with it in real-time while developing, getting immediate feedback without switching context or setting up local simulators.