The Codex feature that works while you sleep

OpenAI's Codex /goal command transforms AI from a turn-based assistant into an autonomous agent capable of working for hours on complex tasks without constant prompting. In a 30-minute episode, Claire Vo demonstrates how the feature eliminated thousands of Sentry errors, cleaned 3,900 emails down to 68, and organized hundreds of Linear tasks during unattended runs. The shift represents a fundamental change in how users interact with AI, moving from babysitting the model to managing it as a self-directed worker.

In this 30-minute episode, I walk through my favorite feature in Codex: the /goal command. I show how Goals transform AI from a turn-based assistant that needs constant ‘what’s next?’ prompting into an autonomous agent that can work for hours on complex, multi-step tasks. I share three real examples: eliminating thousands of Sentry errors, cleaning 3,900 emails down to 68, and organizing hundreds of Linear tasks. Listen or watch on YouTube, Spotify, or Apple Podcasts What you’ll learn: What Goals are and how they differ from standard prompts How I used /goal to eliminate hundreds of error logs in my codebase over a five-hour autonomous run The non-technical use cases that make Goals incredibly powerful: cleaning up 3,900 emails in under four hours and organizing hundreds of project management tasks in Linear How to write effective /goal prompts with measurable outcomes, verification methods, and constraints When not to use Goals and what makes a strong versus weak Goal Why Goals represent a fundamental shift in how we work with AI, from babysitting the model to managing it Brought to you by: Mercury —Radically different banking loved by over 300K entrepreneurs In this episode, we cover: 00:00 https://www.youtube.com/watch?v=2wLJl9A2CnA Introduction 01:50 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=110s What is /goal and when should you use it? 02:45 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=165s The difference between prompts and Goal-based loops 04:06 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=246s Claire’s first five-hour 45-minute autonomous coding task 05:05 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=305s How to manage a Goal lifecycle: view, pause, resume, and clear 06:06 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=366s How to write strong goals: outcomes vs. outputs 07:34 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=454s The six components of effective Goals 08:57 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=537s Example: Reducing P95 checkout latency with /goal 09:36 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=576s Demo: Using /goal to eliminate Sentry errors in ChatPRD 13:18 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=798s Demo: Burning down Vercel API errors 17:28 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=1048s Non-technical use case: Cleaning 3,900 emails with /goal 21:24 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=1284s Demo: Using /goal to clean up Linear project tasks 24:41 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=1481s When not to use /goal 26:10 https://www.youtube.com/watch?v=2wLJl9A2CnA&t=1570s Why /goal changes everything Tools referenced: • Codex: https://openai.com/codex/ https://openai.com/codex/ • Sentry: https://sentry.io/ https://sentry.io/ • Vercel: https://vercel.com/ https://vercel.com/ • Linear: https://linear.app/ https://linear.app/ Other reference: • OpenAI blog post “Using Goals in Codex”: https://developers.openai.com/cookbook/examples/codex/using goals in codex https://developers.openai.com/cookbook/examples/codex/using goals in codex Where to find Claire Vo: ChatPRD: https://www.chatprd.ai/ https://www.chatprd.ai/ Website: https://clairevo.com/ https://clairevo.com/ LinkedIn: https://www.linkedin.com/in/clairevo/ https://www.linkedin.com/in/clairevo/ Production and marketing by https://penname.co/ https://penname.co/ . For inquiries about sponsoring the podcast, email email protected /cdn-cgi/l/email-protection .