Plato Web App

Plato's web application helps you manage and fix Playwright tests across your entire team. It integrates with your CI/CD pipeline to catch and fix failing tests before they block your deployments.

Plato's web app is currently in beta. Join the waitlist to get early access.

How it works

Test Failure Detection
- Monitors your CI pipeline
- Catches failing Playwright tests
- Analyzes failure patterns
Smart Analysis
- Compares page snapshots
- Identifies selector changes
- Determines root causes
Fix Generation
- Suggests targeted fixes
- Validates fix effectiveness
- Creates pull request updates
Team Review
- Review suggested fixes
- Approve or modify changes
- Apply fixes to your codebase

Prerequisites

A Playwright test suite
Node.js 18 or higher
npm or yarn package manager

Package installation

Install the Plato package in your project:

npm install @numen/plato
# or
yarn add @numen/plato

Configure test runner

Update your Playwright configuration to use Plato's test runner:

// playwright.config.ts
import { defineConfig } from "@playwright/test";
import { platoRunner } from "@numen/plato";

export default defineConfig({
  // ...your existing config
  reporter: [
    ["list"],
    [
      "@numen/plato",
      {
        // Plato configuration options
        project: "your-project-name",
        uploadSnapshots: true,
      },
    ],
  ],
});

The Plato reporter automatically captures test context and selectors during test runs. This information is used to generate precise fixes when tests fail.

AI models

Plato offers several AI models to choose from, each with different strengths and characteristics.

Available models

OpenAI models

GPT-4o
- Highest accuracy for complex test fixes
- Best understanding of page context
- Slower response time
- Higher cost per request
GPT-4o Mini
- Good balance of speed and accuracy
- Suitable for most test fixes
- Faster than GPT-4o
- More cost-effective

Anthropic models

Claude 3.5 Sonnet
- Excellent at understanding test context
- Strong selector generation
- Competitive pricing
- Good documentation comprehension

Google models

Gemini 1.5 Flash
- Fastest response times
- Good for simple fixes
- Most cost-effective
- Best for high-volume usage