FROM A RED BUILD TO A CLEAR NEXT STEP
Find the tests
worth investigating.
Use this free tool to turn a Playwright JSON report into a focused failure list and an evidence brief. No payment or AI call is required. Your report stays in this browser.
Start with a test report
Use the standard JSON reporter. Maximum file size: 10 MB.
npx playwright test --reporter=jsonSave its JSON output to a file. Trace ZIPs and HTML reports are not supported.
No report loaded. Nothing is uploaded or stored by this tool.
Run overview
Investigation queue
Take the evidence with you.
Export a plain-text brief with observations and next checks. Error text, logs and attachments are excluded.
Identifiers are omitted by default. Review every export before sharing.
RESOURCE FORGE · DIAGNOSTIC PILOT
See a worked example and compare free versus paid
One failing test. A focused investigation. $49.
One Playwright test in one environment: evidence summary, reproducibility assessment, possible causes with confidence, a correction when supported, and a rerun recipe. One clarification included. A fix is not guaranteed.
Start with scope review, before paying. Email a short problem description; a sanitized evidence brief is optional at this stage. We confirm whether we can accept the case, the delivery date and terms, then assign your case reference. The free automated brief above summarizes your report. The separate $49 service adds an AI-assisted investigation using ChatGPT/Codex, checked against the supplied evidence, for one accepted test failure; buying it is optional.
Check whether your test fits — no payment required
No email app configured? Write to dmteig1213@gmail.com with: what failed, your Playwright version, and local-versus-CI behavior. Start with a short sanitized description; you can send the brief after scope is confirmed.
After scope acceptance only: Pay $49 securely with Square. Enter your approved case reference and purchase quantity one. Payment is verified in Square before work starts. Send your sanitized report by replying to the accepted-case email; results are delivered in that same thread.
Accepting scope-review requests for the pilot. Payment and delivery use a reviewed case workflow; no immediate or unattended turnaround is promised. Turnaround depends on case scope and available capacity; you receive a specific delivery date before deciding whether to pay. No personal human review is promised.
Evidence, not a diagnosis
A retry pass signals a changed outcome, not its cause. Expected failures are labelled accurately. This tool does not inspect traces, run code or certify a release.
Safe submissions
Email only material you own or are authorized to share. Remove credentials, customer personal data, private URLs and sensitive identifiers. Start with plain text and the sanitized brief; do not send raw traces, executables or production access. Diagnostics may use authorized OpenAI tools. Email is not end-to-end encrypted; do not send confidential material that requires a secure portal.
Local tool privacy
This page processes reports in your browser without uploads or storage. Emailing a report is a separate, deliberate sharing step. We count anonymous daily page views and button clicks, without cookies or visitor identifiers. Counts older than 30 days are cleared on the next measurement event. Report contents are never sent for measurement. Do Not Track and Global Privacy Control opt out; hosting providers still handle normal connection data. Counts are not unique people or sales. Only information needed to scope, deliver and record the service is requested. Contact dmteig1213@gmail.com about access or deletion. Financial records may need to be retained separately.