GenAI PMCodex
An AI coding agent or environment mentioned as a place to run AI eval skills. It is also listed as one of the agents that can be compared in a shared environment.
Key Highlights
- Codex appears in the coverage as both a coding agent and a broader execution environment for browser, computer-use, and workflow automation tasks.
- OpenAI reported that Codex accounted for 64% of combined Codex and ChatGPT output tokens as enterprise usage shifted from assistance to execution.
- Practitioners described using Codex for prototyping, eval execution, content operations, QA, inbox workflows, and back-office automation.
- Codex is frequently compared with Claude Code and Cursor, making it relevant for teams evaluating shared agent environments and skills.
- For AI PMs, Codex is most useful as a vehicle for rapid prototyping, workflow automation, and testing governance and eval patterns for AI agents.
Codex
Overview
Codex is an AI coding agent and execution environment that appears across workflows as both a software-building tool and a broader agentic workspace. In the newsletter coverage, it is referenced not just for writing code, but for running eval skills, automating browser and computer tasks, creating prototypes, operating business workflows, and serving as one of the agent environments teams compare alongside tools like Claude Code and Cursor. The aliases and related mentions suggest Codex spans multiple surfaces, including CLI, desktop app, and computer-use/browser-based experiences.For AI Product Managers, Codex matters because it represents the shift from AI as a chat assistant to AI as an operator that can execute multi-step work. The mentions show it being used for prototyping, workflow automation, content operations, eval execution, and enterprise task completion. That makes Codex relevant not only as a developer tool, but as a product surface for testing agent UX, skills, permissions, observability, and cross-tool orchestration inside real business processes.
Key Developments
- 2026-08-08: Codex was mentioned as one of the agent tools used alongside Cursor, Manos, and Claude in workflows connected to Total Recall, a memory layer for recovering past skills, instructions, and chronological agent work.
- 2026-08-11: Instructions were shared for adding fullstackpm CLI skills such as `/docs-audit` and `/wiki-setup` to Codex workspaces, positioning Codex as a configurable environment for reusable PM and documentation workflows.
- 2026-08-13: OpenAI reported that enterprise AI usage is moving from assistance to execution, noting that as of June, Codex produced 64% of combined Codex and ChatGPT output tokens. The report also highlighted sharp weekly-user growth for Codex in legal, sales, recruiting, and marketing, suggesting expanding adoption beyond engineering.
- 2026-08-15: Adi Singh described using an OpenClaw agent with AgentMail to query Codex for relevant newsletter messages and trigger workflows from email threads, showing Codex participating in multi-agent and inbox-driven automations.
- 2026-08-16: Guillermo Rauch predicted that a command for configuring existing coding harnesses, including Claude Code and Codex, could become the default approach for coding AI at scale, emphasizing uptime, model choice, lower costs, observability, and zero data retention.
- 2026-08-17: Ian Silber said OpenAI’s design team uses ChatGPT, Codex, and ChatGPT Work to prototype ideas, test interface decisions, and ship experimental changes quickly. He specifically described using Codex to turn early ideas into prototypes and visual artifacts.
- 2026-08-17: Riley Brown shared a detailed Codex workflow for running a large AI-content business, using Codex skills and sub-agents for YouTube research, transcript extraction, channel scraping, graphics generation with Remotion and Excalidraw, Notion outlining, and thumbnail ideation.
- 2026-08-18: Yana Welinder described using Codex with computer use as part of a workflow that moves from fashion concepts and imagery to research, CAD/prototype work, and a live pre-order website.
- 2026-08-20: Claire Vo shared practical operational uses for Codex browser/Chrome/computer workflows, including accounting, inbox management, Stripe Radar setup, browser QA, security questionnaires, SaaS configuration without APIs, and subscription cancellation.
- 2026-08-25: Peter Yang shared that Shreya and Hamel’s free AI eval skills from the `ai-evals-course` GitHub repository can be run in Codex, reinforcing Codex as an environment for evaluation workflows in addition to product execution.
Relevance to AI PMs
1. Prototype faster across product and design workflows. Codex is repeatedly mentioned as a way to turn rough ideas into working prototypes, artifacts, and experiments. AI PMs can use it to validate UX concepts, test flows before engineering commitment, and compress the cycle from concept to stakeholder feedback.2. Operationalize agentic workflows beyond coding. The examples span QA, inbox work, browser automation, research, content production, and back-office tasks. For AI PMs, this makes Codex useful for identifying high-ROI workflows where an agent can execute real work, especially when APIs are incomplete and browser/computer use is required.
3. Design eval and governance patterns for AI systems. Codex is cited as a place to run eval skills and as part of enterprise usage trends toward execution. PMs can use it to test reusable skills, define permissions, measure outcomes, and compare agent performance across environments like Claude Code and Cursor.
Related
- OpenAI: Codex is closely associated with OpenAI and appears in OpenAI reporting on enterprise adoption and execution-heavy usage.
- ChatGPT / ChatGPT Work / ChatGPT Pro: These are adjacent OpenAI product surfaces often mentioned alongside Codex, especially for prototyping, summarization, and cloud-based work.
- Claude Code: Frequently referenced as a comparable coding-agent environment and an alternative place to run shared skills and eval workflows.
- Cursor: Another coding and agent workspace commonly compared with Codex in shared tooling setups.
- Codex CLI / codex-cli / Codex desktop app / codex-app: These aliases suggest Codex exists across multiple interfaces, from command-line workflows to desktop and app-based experiences.
- Agent-working-protocol / OpenTelemetry / openai-compliance-platform: These related entities connect to themes important for enterprise agent deployment, including interoperability, observability, and governance.
- GitHub, Slack, Gmail, Stripe, Notion, Linear: These integrations and adjacent tools show the kinds of systems Codex is discussed alongside in real operator workflows.
Newsletter Mentions (96)
“Peter Yang shared ai-evals-course’s GitHub repository containing free AI eval skills attributed to Shreya and Hamel, which can be run in Claude Code or Codex.”
GenAI PM Daily August 25, 2026 GenAI PM Daily 🎧 Listen to this brief 3 min listen Today's top 19 insights for PM Builders, ranked by relevance from Blogs, YouTube, and LinkedIn. GPT-5.6 in Kiro advances developer price-performance #1 📝 OpenAI News Advancing price-performance for developers with GPT‑5.6 in Kiro - Announces availability of GPT‑5.6 in Kiro to improve price-performance for developers, enabling more cost-effective and performant model access for applications. #5 𝕏 Peter Yang shared ai-evals-course’s GitHub repository containing free AI eval skills attributed to Shreya and Hamel, which can be run in Claude Code or Codex.
“Claire Vo shared how she uses Codex browser/Chrome/computer for operational tasks including accounting, inbox management, Stripe Radar configuration, browser-based QA, security questionnaires, SaaS setup when an API is unavailable, and subscription cancellation.”
GenAI PM Daily August 20, 2026 GenAI PM Daily 🎧 Listen to this brief 3 min listen Today's top 20 insights for PM Builders, ranked by relevance from Blogs, X, YouTube, and LinkedIn. OpenAI announces Zero Data Retention for frontier models #1 📝 OpenAI News Offering Zero Data Retention for frontier models - OpenAI announces offering zero data retention for frontier models, committing to not retain user data for those models and clarifying how this impacts customers and data handling. The post outlines the company's privacy-focused approach for frontier model interactions. Also covered by: @OpenAI , @OpenAI , @Sam Altman #2 𝕏 Cursor announced that it can now monitor pull requests, watch a Slack thread, and run scheduled tasks. Cloud agents automatically subscribe to pull requests they create and drive them to completion. #3 𝕏 Mustafa Suleyman announced that MAI-Image-2.5 is ranked #1 on the Artificial Analysis leaderboard for image editing. #4 𝕏 Logan Kilpatrick announced that Google AI Studio now supports GitHub repository imports and bi-directional push/pull synchronization. A new UI also supports force pushes and merges. #5 𝕏 Qwen shared that Qwen3.8-27B ranked as the #1 open-weight model on Harvey’s Legal Agent benchmark, describing it as capable of professional tasks while remaining small enough to run locally. #6 𝕏 NVIDIA shared that NVIDIA cuOpt, its open-source solver, is the fastest open-source solver on Hans Mittelmann benchmarks across three optimization problem classes. #7 𝕏 Results from benchmarks of 300+ NVIDIA verified skills on real tasks showed that using skills improved correctness by 41 points, effectiveness by 39 points, and efficiency by 35 points. SkillEvaluator is open source for testing skills before shipping. #8 𝕏 Philipp Schmid shared that Gemini 3.7 Flash ranked first on Artificial Analysis’s new AA-AnalystAgent, which covers 80 real-world quantitative analysis tasks across 14 business and scientific domains. #9 𝕏 Claire Vo shared how she uses Codex browser/Chrome/computer for operational tasks including accounting, inbox management, Stripe Radar configuration, browser-based QA, security questionnaires, SaaS setup when an API is unavailable, and subscription cancellation.
“Yana Welinder uses ChatGPT Images 2.0, Codex with computer use, CLO 3D, deep research, Superhuman, GitHub, and Stripe to take Yana Bana garments from hand-drawn sketches or visual references through fashion images, manufacturing research, CAD/prototype work, and a live pre-order website.”
GenAI PM Daily August 18, 2026 GenAI PM Daily 🎧 Listen to this brief 3 min listen Today's top 20 insights for PM Builders, ranked by relevance from X, YouTube, LinkedIn, and Blogs. Cursor releases Origin, its integrated code hosting platform #1 𝕏 Cursor released Origin, its code hosting platform, with deep Cursor integration and repository syncing from GitHub. Cursor describes Origin as fast and easy to use. Also covered by: @Cursor , @Guillermo Rauch #2 𝕏 Philipp Schmid demonstrated Gemini 3.7 Flash using his Android emulator via ADB for the task “Play 1 round of Wordle.” He said its latency and visual reasoning make it exceptionally good for multimodal agentic use cases such as mobile control and Computer Use. #5 ▶️ How I use ChatGPT to run my fashion business How I AI Podcast Yana Welinder uses ChatGPT Images 2.0, Codex with computer use, CLO 3D, deep research, Superhuman, GitHub, and Stripe to take Yana Bana garments from hand-drawn sketches or visual references through fashion images, manufacturing research, CAD/prototype work, and a live pre-order website.
“Ian Silber describes how OpenAI’s product-design team uses ChatGPT, Codex, and ChatGPT Work to prototype ideas, test durable interface decisions, and ship experimental product changes quickly.”
#5 ▶️ OpenAI’s Head of Design: This is the best time in history to be a designer | Ian Silber Lennys Podcast Ian Silber describes how OpenAI’s product-design team uses ChatGPT, Codex, and ChatGPT Work to prototype ideas, test durable interface decisions, and ship experimental product changes quickly. Ian Silber said engineers have sometimes increased productivity by 10x to 100x with AI, while design work still requires repeated feedback loops, including trying ideas, discarding them, and gathering user or internal feedback. For some ChatGPT features, OpenAI tries about 100 options, discards 99, and ships one; for other work, teams use a “building in public” approach, take large swings, and learn from rapid feedback. Ian Silber uses ChatGPT Work, which runs in the cloud, and Codex to turn early ideas into prototypes or visual artifacts, including from a phone when he is away from a computer; he also uses AI to summarize Slack follow-ups and prepare context for meetings and recruiting.
“Riley Brown uses Codex skills to research YouTube videos, generate Remotion graphics and Excalidraw diagrams, draft Notion content outlines, and build Paper-based YouTube thumbnail variations for his 1.5M+ follower AI-content business. His YouTube researcher skill uses the Supadata API to pull a full transcript from a YouTube video in about one second; Codex sub-agents can scrape an entire channel in about 30 seconds.”
#3 ▶️ How I Run My 1.5M+ Follower Content Business With Codex | Riley Brown Peter Yang Riley Brown uses Codex skills to research YouTube videos, generate Remotion graphics and Excalidraw diagrams, draft Notion content outlines, and build Paper-based YouTube thumbnail variations for his 1.5M+ follower AI-content business. His YouTube researcher skill uses the Supadata API to pull a full transcript from a YouTube video in about one second; Codex sub-agents can scrape an entire channel in about 30 seconds. He chains an “internet image puller” skill using SerpAPI and Google Images with a Remotion best-practices skill: the agent pulls relevant logos from a video transcript, creates branded graphics, and exports them as overlays for edited video. For thumbnails, Codex scrapes high-performing thumbnails—including Alex Hormozi and Dan Martell examples—into Paper, then Riley Brown uses Paper’s built-in AI image generation to replace the original subject with his own image and iterates on details such as face smoothing, outline glow, text, and colors.
“in Guillermo Rauch predicts that an unnamed command for configuring existing coding harnesses, including Claude Code and Codex, will become the default way to use coding AI at scale.”
#3 in Guillermo Rauch predicts that an unnamed command for configuring existing coding harnesses, including Claude Code and Codex, will become the default way to use coding AI at scale. He highlights uptime, model choice, lower costs, observability, and ZDR.
“Adi Singh uses an OpenClaw agent with AgentMail to subscribe to newsletters, query Codex for relevant messages, forward client deals and documents for Claude to parse, and trigger workflows by CC’ing an agent in an email thread via webhooks.”
#8 ▶️ AgentMail Founder Adi Singh on Email for Agents SyntaxGTM Adi Singh uses an OpenClaw agent with AgentMail to subscribe to newsletters, query Codex for relevant messages, forward client deals and documents for Claude to parse, and trigger workflows by CC’ing an agent in an email thread via webhooks.
“OpenAI published two reports showing enterprise AI is shifting from assistance to execution: as of June Codex produced 64% of combined Codex and ChatGPT output tokens, and frontier firms (top 10% by usage) now generate 8.3× as many output tokens per active user as typical firms (up from 2.6× in January).”
#8 📝 OpenAI News How enterprises put AI to work - OpenAI published two reports showing enterprise AI is shifting from assistance to execution: as of June Codex produced 64% of combined Codex and ChatGPT output tokens, and frontier firms (top 10% by usage) now generate 8.3× as many output tokens per active user as typical firms (up from 2.6× in January). Frontier adopters use advanced capabilities far more—21% of weekly active users at frontier firms use Plugins (vs. 9% at typical firms) and 19% use skills (vs. 3%)—Codex weekly users grew 108× in legal, 41× in sales, 41× in recruiting, and 26× in marketing (vs. 5× in engineering), and OpenAI recommends connecting agents to company context/tools, strong permissions and governance, and shared workflows to scale adoption.
“He also shared instructions for adding the fullstackpm CLI’s /docs-audit and /wiki-setup skills to Claude Code, Codex, or Cursor workspaces.”
He also shared instructions for adding the fullstackpm CLI’s /docs-audit and /wiki-setup skills to Claude Code, Codex, or Cursor workspaces.
“The speaker uses Codex, Cursor, Manos, and Claude.”
#7 ▶️ Total Recall: The Invisible Memory Layer for AI Agents SyntaxGTM Total Recall runs as a background memory layer for AI agents, storing session instructions, decisions, and work history so queries can recover workflows, skills, and chronological project changes across coding sessions. A query asking “Can you recall how I upload videos to YouTube?” retrieved an older YouTube upload skill, including its Drive-upload work, an Auphonic update, and consolidation of an archived skill into a newer upload-to-YouTube skill from roughly three months earlier. Total Recall can search all agent sessions and produce a date-based, Git-log-style list of changes for client reporting, including work performed without a GitHub repository or properly maintained commits; the speaker uses Codex, Cursor, Manos, and Claude. For the vague request to recall the latest three skills, Total Recall issued 10 queries and identified an Agent Mail inbox workflow, a sponsor CRM skill for five to ten daily sponsor requests, and a skill related to an autonomous AI marketing-consultancy/product-launch project.
Related
An AI coding assistant environment used for running evaluation skills and agentic workflows. In this issue it is mentioned as a runtime for ai-evals-course material and as an agent in an OpenRouter-like system.
An AI company best known for Claude. It is referenced implicitly through Claude’s memory and Cowork features.
An AI company building frontier models, ChatGPT, and custom inference hardware. Here it is discussed for Jalapeño and ChatGPT Business Premium Seats.
Anthropic’s assistant, discussed here for shared memory across chat and Cowork. The feature is relevant to PMs because it enables cross-task context reuse and user-controlled memory.
An AI coding tool referenced as providing data used to evaluate Grok 4.6. It is also named later as a target environment for running AI eval skills.
Founder and CEO of Vercel, cited here announcing Run SDK and Vercel Connect. He is influential in developer tooling and AI app infrastructure.
A creator/curator in the AI PM space who shared the ai-evals-course repository. He is mentioned as a source for practical AI eval resources.
A prominent AI blogger and commentator referenced in connection with an article on token reselling and fraud. He is cited as the source of the newsletter item discussing the marketplace and API-key abuse.
A model and dataset platform referenced as the source of the supported model used by TensorRT Model Connect. Important for PMs working with open model ecosystems and evaluation artifacts.
Product and business commentator who reacted to Ethan Mollick’s post about AI changing work roles. Included here because he is discussing organizational and role boundaries in the AI era.
NVIDIA’s AI organization, referenced for model benchmarking and rankings. The newsletter notes its Nemotron model performance in PinchBench and OpenClaw tests.
A standardized agent test suite referenced for model evaluation. The newsletter cites success rates on OpenClaw as part of the Nemotron benchmark result.
OpenAI’s conversational AI product used by the design team to prototype ideas and test interface decisions. Here it is also part of a rapid experimentation workflow.
AI researcher and educator known for clear explanations of model sampling and watermarking. Here he explains watermarking in terms of top-p/top-k selection.
An operator or product thinker who raised concerns about data indexing, connector visibility, prompt injection, and evaluation quality. Her comment focuses on trust, deletion, and user-empathetic system design.
CEO of OpenAI and a key public figure in frontier AI product and policy announcements.
A major AI infrastructure company developing hardware and software for training and serving models. In this newsletter it appears in the context of Dynamo, GLM-5.2 testing, and open model routing.
Person who shared an agent skill in the treg repository. Relevant to PMs because it showcases community distribution of reusable agent behaviors.
A prominent AI researcher and educator, quoted here on compilation and IR design in relation to PyTorch and microgpt-like specifications. He is often cited for deep technical product and model architecture insights.
An autonomous coding agent used by a solo founder to manage engineering work and PR workflows. The newsletter highlights extensive threaded usage, playbooks, and review automation around it.
A CRM and software company whose APIs were rated highly in a product API access grading exercise. The newsletter highlights its documentation and UI as useful for prototyping workflows.
An OpenAI product leader mentioned as the user of Codex for product work. He is described as using AI to synthesize feedback, prototype interfaces, and automate operational workflows.
There's An AI For That is an AI discovery platform that curates tools and use cases. Here it is cited for emphasizing the importance of context in agent behavior and introducing HydraDB.
Henry Shi is a technical staff member at Anthropic Labs and co-runner of the AI Product Management Certification. He is described as a former co-founder of Super.com.
A model used as an automated judge in Claire Vo’s benchmark. It contributes 30% of the scoring alongside her manual evaluation.
A workplace messaging platform used here as an operational surface for AI agents. PMs may care because agent integrations increasingly extend into team communication workflows.
A payments and commerce infrastructure tool used to support pre-orders in the fashion-business workflow described. Relevant for AI PMs building monetization and checkout flows.
Linear is a product and issue-tracking company whose team shared practical guidance for building production agents.
A collaborative design platform referenced as an example of broad enterprise SaaS that may remain resilient in the AI era. It is contrasted with niche single-purpose products.
OpenAI's coding agent system used here to build NVIDIA AI's TensorRT Model Connect and also referenced as a benchmarked assistant in connector support comparisons. Relevant to PMs considering AI-assisted software engineering.
A frontier model release referenced as improving price-performance for developers. It is discussed as being available in Kiro for more cost-effective application development.
An embeddable assistant capable of streaming answers, rendering UI, and acting within applications.
Cloud Code appears to be a coding agent or coding workflow used to generate launch videos from websites. The newsletter describes it as working with Fable 5 and HyperFrames.
A software development platform used here as the source and sync target for repositories. It is central to AI coding workflows, plugin distribution, and agent automation.
A workspace and note-taking tool used here to store research outputs as cards. In this workflow it supports agent-generated content operations.
OpenAI product leader/executive who publicly praised GPT-5.2 in the newsletter. Useful context for AI PMs tracking product and model reception.
Co-founder and CEO of Databricks, mentioned here in connection with AI Extract and SQL-callable PDF extraction. He is highlighted as discussing accuracy and cost improvements for document extraction workflows.
A GPT model variant used here for scientific reasoning and agentic chemistry experimentation. The newsletter frames it as a model capable of proposing experimental improvements and driving benchmarked workflows.
xAI’s AI assistant/model family, mentioned here as part of a set of subagents coordinated through tmux. It is relevant to PMs tracking competing agent products and workflows.
A coding tool or interface used to connect Kimi K3 to Polymarket data in a trading workflow. It functions as the orchestration layer for market analysis and execution.
A tool for generating video graphics and programmatic video content. Here it is used within a Codex-powered workflow to create branded overlays.
A cloud-run version of ChatGPT used here to prototype ideas and create artifacts away from a computer. It is presented as a practical assistant for mobile and cloud-based workflows.
A coding and research tool used here for optimizing order execution latency. Relevant to PMs as part of an AI-assisted quantitative workflow.
Google’s hidden watermarking technology for AI-generated content across images, video, audio, and text. It is relevant to PMs working on content provenance, trust, and detection.
A creator and operator mentioned in a workflow demo using GPT-5.6, Codex Desktop, and plugins. He appears in the context of automating communications and building a SaaS prototype.
An AI coding IDE and collaboration platform focused on tasks, agent sessions, artifacts, and workflows from questions through implementation. It is positioned as a productivity platform for engineers using AI APIs and agents.
A data cloud platform used as the data source for AI-generated dashboards in this newsletter. It is paired with v0 and Next.js for frontend generation.
OpenAI’s coding-focused model/release highlighted for benchmark performance, steerability, and speed improvements. The newsletter frames it as a strong coding agent option with multiple benchmark scores.
A production tool used with Adobe Premiere in an AI-assisted ad creative workflow. It helps automate or accelerate post-production for marketing content.
Google’s email product, referenced as a connector in Google AI Studio.
An image-generation capability used here for generating product photos and fashion imagery. Relevant for PMs exploring multimodal content creation workflows.
AWS’s managed foundation-model platform for deploying and accessing AI models. In this newsletter it is the distribution channel for OpenAI’s Daybreak cybersecurity models.
OpenAI's chat model optimized for more engaging conversation, better intent understanding, and improved handling of complex constraints. It is described as rolling out to paid users first and then free users.
Enterprise software company mentioned as a customer in a Claude Code migration story. The newsletter highlights a major reduction in migration time and high test coverage.
Google’s command-line interface for working with Gemini in developer workflows. It is mentioned as a compatible tool alongside agent skills in antigravity.
A paid ChatGPT subscription tier with expanded model access and higher usage limits. For AI PMs, this is a packaging and monetization lever that affects power users and workflow depth.
OpenAI leader and product/engineering voice associated here with confirming Codex’s unification with the main model. The newsletter cites him via Simon Willison’s note.
Anthropic’s Claude model used locally in Paperclip’s agent orchestration demo. It is used for task execution, company simulation, and coding workflows.
OpenTelemetry is an observability standard for traces, logs, and metrics. The newsletter mentions Codex exporting agent-aware telemetry through it for auditing and monitoring.
A Python-derived clone created from leaked Claude Code TypeScript. It is described as a fast-growing GitHub repo.
Stay updated on Codex
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free