4Hacker News (AI-filtered, score >= 200)·
OpenAI Codex CLI lacks mechanism to exclude sensitive files from context
A GitHub issue on the OpenAI Codex repository highlights an unresolved feature request for excluding sensitive files from the agent's context window, garnering 166 HN points and 110 comments. The absence of this capability raises practical security and privacy concerns for developers using Codex in codebases containing credentials, secrets, or proprietary data. Community engagement suggests this is a meaningful gap in the tool's enterprise readiness.
Related guides (3)
Related events (8)
Running Codex Safely at OpenAI
OpenAI published a blog post describing the security architecture used to run Codex as a coding agent internally, covering sandboxing, human approval workflows, network policies, and agent-native telemetry. The post is aimed at supporting enterprise adoption of coding agents by demonstrating safe and compliant deployment patterns. It provides operational detail on how OpenAI itself governs agentic code execution in production.
OpenAI reduces Codex CLI model context window from 372k to 272k tokens
A pull request in the OpenAI Codex repository reduces the model context size from 372k to 272k tokens, a ~27% reduction. The change was noticed by the Hacker News community and generated moderate discussion. Context window reductions in coding agents are notable because they affect the size of codebases and task horizons the tool can handle in a single session.
OpenAI Codex adds Team Config for shared configuration across repositories and machines
OpenAI's Codex now supports a Team Config system that allows teams to standardize configuration across repositories and machines by sharing config files, command-control rules, and reusable workflow skills. Configuration is loaded from a layered hierarchy of .codex/ folders spanning working directory, parent directories, repo root, user home, and system locations, with higher-precedence locations overriding lower ones. Admins can enforce hard constraints via a requirements.toml file that overrides all other defaults. This is a tooling update aimed at enterprise and team-scale deployment of Codex.
Codex Security: now in research preview
OpenAI has launched Codex Security in research preview, an AI-powered application security agent. It analyzes project context to detect, validate, and patch complex vulnerabilities with the goal of higher confidence and reduced false-positive noise compared to traditional tools. The product extends OpenAI's Codex brand into the security domain.
OpenAI adds Codex access tokens for automated CI/script workflows in ChatGPT Enterprise
OpenAI has added support for Codex access tokens in ChatGPT Enterprise, allowing workspace admins to permit members to authenticate Codex for non-interactive, automated workflows. Tokens enable Codex to run from scripts, schedulers, and private CI runners while maintaining ChatGPT workspace identity. This extends Codex's reach into developer automation pipelines beyond interactive use.
OpenAI Codex gains goal mode GA, Appshots context, and remote locked computer use
OpenAI shipped a batch of Codex updates including general availability of goal mode across the app, IDE extension, and CLI, allowing users to define outcomes and success criteria for long-running autonomous tasks. Appshots on macOS let users attach app windows to Codex threads for richer visual context without verbose setup prompts. Locked computer use enables eligible Mac users to keep Codex running remotely after the machine locks, extending the agentic reach of the system. Browser use also received annotation, asset extraction, and reliability improvements.
OpenAI Codex app gains integrated terminal context reading
OpenAI's Codex app now reads the integrated terminal within the current thread, enabling it to inspect running development server status or failed build output during coding sessions. The update also includes unspecified performance improvements and bug fixes. This is a targeted agentic capability improvement that tightens the feedback loop between Codex and the developer's live environment.
OpenAI Codex for Business adds coworker invitations, Developer mode, and app access controls
OpenAI released several Codex updates for Business users, including coworker invitation flows with referral bonuses, a Developer mode enabling Chrome DevTools Protocol (CDP) access for debugging in the Codex browser, and per-app access controls for Computer Use on Windows. Additional changes include an /init command for generating AGENTS.md scaffolds, clearer usage-limit error messages, and an Unread chats section in the command menu. These updates expand Codex's enterprise administration controls and developer tooling depth.