Run browser-based QA across primary and adversarial paths, producing reproducible bugs instead of a vague test summary.
06:51 to 12:26: Claire uses Codex browser automation to exercise a web app beyond the happy path and return a detailed record of the tested flows and failures.
A QA report with passed paths, reproducible failures, screenshots, and affected environment details.
OpenAI conversational AI
Step by step
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
5 steps
Download and install an AI desktop app with browser control, such as the ChatGPT or Claude desktop app. You will also need its corresponding Chrome extension to allow the AI to see and control your browser.
Install the approved desktop app and browser extension, then confirm the test profile and environment.
Open the web application you want to test in the AI app's browser. This can be a live URL or a local development environment running on your machine.
Open [test URL], sign in as [role], and confirm the seeded records are present.
Instruct the AI to test the application using a directive prompt. Specify the user flow to test, the criteria to evaluate, such as usability and mobile responsiveness, and the desired output format for the bug report.
Test these journeys: [list]. Include happy, invalid, empty, repeated, mobile, and permission cases. Do not perform destructive production actions.
Watch as the AI agent navigates your site. It will methodically click buttons, test failure states like submitting a form with a required field left blank, and change viewport sizes to simulate different devices.
Record each action, screenshot failures, and stop if authentication or environment differs from the brief.
Review the bug report generated by the AI. A typical output is a spreadsheet that lists each issue with a description, steps to reproduce, and the corresponding screenshot for context.
Return a bug table with severity, URL, role, data, steps, expected, actual, and evidence, plus passed checks.
Turn an idea into a PRD, user stories, and a plan.
After the steps
How to recover when the loop fails and where human judgment helps.
Recover
Keep building
Log your workouts effortlessly without ever opening a spreadsheet. This workflow uses the Claude mobile app to parse a simple voice note and automatically update your workout tracking spreadsheet in the background.
Create a custom email management interface within Claude to read, triage, and draft replies without opening the Gmail tab. This workflow helps you reclaim focus and compound your learnings by training your AI on every interaction.
Build a fully automated system in Claude to generate personalized, interactive HTML proposals and client onboarding materials, saving hours on administrative work and creating a high-touch experience.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.