Playwright Test Debugger

Overview

This agent acts as an expert Playwright test debugging specialist. Its primary function is to move beyond simply reporting a failure; it systematically diagnoses why a test fails or behaves flakily by executing multiple diagnostic runs and analyzing detailed traces.

It follows a structured, multi-step protocol designed to isolate the root cause—whether it's timing issues, state leakage, or environmental discrepancies.

Capabilities

• Multi-Run Execution: Executes tests under various conditions (--repeat-each, --workers=1) to expose intermittent failures.
• Trace Analysis: Reads and interprets Playwright traces for network errors, visibility issues, and timing anomalies.
• Failure Classification: Categorizes the failure into distinct types: Timing/Async, Test Isolation, Environment, or Infrastructure.
• Root Cause Identification: Provides actionable advice on common causes like missing await calls, state pollution, or race conditions.

Example Use Cases

1. Intermittent Failure: A test passes locally but fails randomly in CI. The agent will check for environment differences and potential state leaks.
2. Timeout Errors: When an element locator times out intermittently, the agent can help determine if it's due to missing waits or slow network dependencies.
3. Debugging Flakiness: If a test only fails when run as part of a large suite, this agent helps pinpoint if the failure is caused by one test polluting the state for another.