Playwright Suite Audit

The problem

Your end-to-end suite is red for reasons nobody can explain, CI takes too long, and the team re-runs jobs until they pass. I review the suite and the pipeline and hand you a prioritised fix list.

For: Engineering managers and QA leads whose Playwright suite is slow, flaky or not trusted

What I do

A fixed-scope review of the suite, its config and the CI pipeline, plus the flaky-test history of your last 30 days of runs. You get a written report where each finding has a severity, file and line evidence, the fix and an effort estimate, and a debrief call to walk through it.

How to start

1. 1Fit call

   Free 30 min

   A short call to look at your setup. If an audit is not the right fix, I say so on the call.

   • 30-minute video call
   • No preparation needed

2. 2Paid diagnostic

   €390 one-time

   For teams not ready to share code yet. Conversation and documents only, no repo access. The fee is credited in full against the audit if you book it afterwards.

   • 90-minute session
   • Pre-session questionnaire
   • Written action plan within 48 hours
   • One follow-up call within 30 days

3. 3Playwright Suite Audit

   €1,650 flat fee, confirmed in a written quote

   Report within 10 business days of repo access.

   • Kickoff call (45 min) to agree the scope and the pain points
   • Read-only review of the Playwright suite, config and CI pipeline
   • Flaky test analysis from your last 30 days of CI runs
   • Written report: findings ranked by impact and effort, with code-level examples
   • Debrief call (60 min) to walk through the report

4. 4Fix Sprint

   Quoted after the debrief

   Optional follow-on: I implement the top findings from a delivered audit as pull requests on your repository.

   • Pull requests for the findings we agree on
   • Before and after measurement in CI (flaky-test count, job duration)
   • Short handover note
   • Time-boxed; only findings from a delivered audit

Proof

• A full sample report on a public suite: 10 findings (3 high, 4 medium, 3 low)

  Each finding links to the file and line it is about, with the fix and an effort estimate.

  Source: Sample report, 2026-10-01

• The audited demo suite: 8 tests on 3 browser engines, 24 of 24 passing in CI

  Page objects, fixtures, test-id locators and no fixed waits.

  Source: CI run on main, 2026-10-05

• Finding H1 from the sample report has since been fixed; H2 is still open

  A weekly scheduled CI run now keeps the green badge current (commit d0221dc).

  Source: playwright-e2e-demo history

• Open-source locator fallback engine: when a test id is removed, it recovers the element by role and reports the match as healed

  11 automated tests passing; published on PyPI as self-healing-browser-mcp 0.1.0.

  Source: Local demo run, 2026-10-10

Terms in short

The report is accepted on delivery unless you raise concrete gaps against the agreed scope within 5 business days; I then correct them once at no cost. Your code, CI logs and the report stay confidential and are never used as a public sample without written permission.