AI-Assisted Content Generation and Documentation

7/20/2026

Introduces custom repository rules for Codex Code Review, enabling developers to define specific checks in `AGENTS.md` to guide code reviews. This feature aims to address common review bottlenecks by encoding team knowledge and invariants, such as API contract preservation and data boundaries, directly into the review process. The post details how these rules are applied, provides examples of rule writing and their impact on review findings, and discusses best practices for creating effective and scoped rules. It highlights the improvement in identifying intended violations and reducing unnecessary findings.

3/20/2026

This post details how GPT-5.4 has been trained with a focus on improved UI capabilities and image usage for frontend development. It introduces practical techniques for steering the model towards desired designs, including leveraging image search and generation tools, developing more complete and functionally sound apps, and utilizing computer use and verification tools like Playwright for iterative inspection and refinement. The post also provides specific prompt guidance and introduces a dedicated 'frontend-skill' to enforce design principles and improve the quality of generated interfaces.

2/23/2026

This post details an experiment demonstrating GPT-5.3-Codex's capability for long-horizon agentic coding tasks, specifically building a design tool over 25 hours. It highlights the importance of the agent loop (plan, edit, run tools, observe, repair, update docs) and the use of durable project memory through markdown files (spec, plan, implement, documentation) for maintaining coherence and task completion. It also introduces the concept of milestone-based verification and repair, and discusses the role of the Codex app in providing structured context and guardrails for such tasks.

2/4/2026

This post details practical lessons learned from building real-world ChatGPT Apps, focusing on the unique challenges of integrating AI models with user interfaces and traditional application patterns. It introduces concepts like 'context asymmetry,' the importance of explicit information sharing between UI and model, and the need for 'lazy-loading' to be reversed in agentic environments. The post also covers UI design considerations for multi-modal displays, consistency within the ChatGPT ecosystem, language-first filtering, and leveraging files for richer interactions. Finally, it touches on production deployment aspects like CSPs. This post significantly expands the understanding of how to effectively build and deploy AI-powered applications within the ChatGPT ecosystem, building upon the initial concept of ChatGPT Apps as a new capability.

1/11/2026

This post details the integration of OpenAI's Codex CLI with JetBrains IDEs using the Model Context Protocol (MCP). This integration allows Codex to leverage IDE context, including file problems (analysis of errors and warnings via IntelliJ inspections) and the execution of predefined run configurations (unit tests, linters, formatters). This enables Codex to receive immediate feedback on code quality and correctness, similar to human developers, leading to faster iteration cycles and higher-quality code suggestions. The post highlights practical examples of catching compilation errors and the benefits of automated testing and formatting.

11/24/2025

This post provides practical guidance on designing and building ChatGPT Apps, focusing on translating product strengths into well-defined capabilities for the model. It introduces the 'Know', 'Do', 'Show' framework for app value proposition and outlines a process for selecting and defining app operations based on core jobs-to-be-done. It also details design considerations for conversational integration and discovery, including handling vague and specific user intents.

10/27/2025

This post details how Dagster Labs leverages OpenAI's Codex to enhance their technical documentation and content creation processes. It highlights the use of Codex for: 1. Accelerating documentation by using structured CONTRIBUTING.md files as input. 2. Acting as a code explainer for pull requests and codebase sections. 3. Translating content across different mediums (e.g., tutorials to video transcripts). 4. Evaluating documentation coverage by generating code from documentation and testing its functionality. The post emphasizes the importance of context and structure, particularly within monorepos, for effective AI-assisted content generation.

10/10/2025

This post details the extensive application of Codex in building and executing various components for OpenAI DevDay 2025. It showcases how Codex was used for controlling cameras and lighting systems via the VISCA protocol, developing a beat pad interface for a keynote demo, generating multiple game implementations for an arcade game, migrating a Streamlit application to a Next.js/FastAPI architecture, creating UI from sketches and generating evaluation code for SVG generation, refactoring codebases, and producing Mermaid diagrams for system explanation. It also highlights Codex's role in accelerating documentation organization and creating a new 404 page.