Ep 884: Codex Record and Replay: How to Teach an Agent Once Your Most Time-Consuming Workflows (Replay)
October 19, 2026
AI Summary
5 min readCodex Record and Replay: Teaching an AI Agent by Showing, Not Telling
OpenAI quietly shipped a feature that lets you teach an AI agent by simply doing your work in front of it. The host of the Everyday AI podcast calls it a preview of where the entire interface is heading: "Prompting is kind of losing to demonstration." Codex Record and Replay, currently available only on Mac, records everything you do on your screen—every click, every highlight, every paste—and turns that sequence into an editable skill file that the agent can then run autonomously. The host demonstrates it live, unscripted, building a daily workflow that pulls the five most-clicked stories from his newsletter, copies their text into NotebookLM, and generates a debate-style audio overview. The whole thing takes about three minutes to record and produces a reusable automation.
What Actually Changed Inside Codex
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of Everyday AI Podcast – An AI and ChatGPT Podcast
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 (00:01) **Why Record and Replay Matters** - The host introduces the new OpenAI/Codex feature and explains why showing your work is replacing prompt engineering.
- 2 (03:03) **What You'll Learn in This Episode** - A quick roadmap of the four key takeaways.
- 3 (04:26) **Live Demo Setup: Installing Record and Replay** - The host walks through prerequisites and security considerations.
- 4 (07:00) **The Demo Workflow: From Newsletter to Daily Audio Brief** - The host shows a real use case: automating a daily research brief.
- 5 (11:55) **Live Recording in Action** - The host demonstrates the screen recording process and its quirks.
- 6 (16:23) **Stopping the Recording and Building the Skill** - The host clicks "Done" and explains what happens next.
- 7 (18:35) **The Real Business Artifact: The Skill .md File** - The host explains why the skill file is more important than the demo itself.
+ Full timestamped outline available in the app
Show Notes
Ever wish your agent would just watch you work and copy you? 👯
Oh wait, that's exactly what OpenAI released.
The new 'Record and Replay' feature in Codex allows you to simply record anything you do and turn it into a repeatable workflow for your agents.
Join us LIVE as we show you how. ⏺️
Ep 884: Codex Record and Replay: How to Teach an Agent Once Your Most Time-Consuming Workflows (Replay)
Newsletter: Sign up for our free daily newsletter
More on this Episode: Episode Page
Today's Episode on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.
Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Website: YourEverydayAI.com
Email The Show: [email protected]
Connect with Jordan on LinkedIn
Topics Covered in This Episode:
- Codex Record and Replay Feature Overview
- Transition from Prompting to Demonstration
- Skill Markdown Files as Business Artifacts
- Codex vs Claude Desktop Agent Comparison
- Workflow Recording and Security Best Practices
- Step-by-Step Codex Record and Replay Demo
- Video Understanding Capability in Codex
- Editing and Sharing Skills Across Platforms
- Automating Multi-App, Cross-System Workflows
- Validation, Scheduling, and Automation of Skills
Timestamps:
00:00 OpenAI's record and replay feature
04:41 Preparing for screen sharing
08:56 Managing Small Tasks Efficiently
10:03 Automating top story analysis
15:43 Debating AI business strategies
18:29 Codec's video ingest feature
22:01 Practicing low stakes automation
24:57 Using skills and plugins efficiently
26:19 Using Chattopty for workflow automation
Keywords:
codex record and replay, OpenAI record and replay, record and replay feature, codex workflow automation, proactive AI agents, skill.md file, skill markdown file, editable skill files, demonstration over prompting, desktop AI agents, Mac AI features, video ingest
More from this podcast
Everyday AI Podcast – An AI and ChatGPT Podcast →