Codex Updates & Release Notes

Chat while you work

Keep ChatGPT close while you work in other apps. In the ChatGPT desktop app on macOS and Windows, you can now type a quick chat from the floating Pets controls and press Enter to send it. Use @ to add context and $ to choose a skill. Select the bell icon to follow progress in your threads, then select a thread to open the full conversation.

Choose a companion or show just the controls in Settings > Pets. Press Option+Space on macOS or Windows+Alt+P on Windows to show the controls and focus Quick Chat. Pressing the shortcut again keeps Quick Chat open. You can change Show pet in Settings > Keyboard shortcuts. To hide the controls, select Hide pet in the command menu or Settings > Pets.

On macOS, you can also send an appshot to the floating controls. Keep the main ChatGPT window in the background and choose Automatic for Appshot destination.

Learn more about Pets.

Share an app window on Windows

Appshots are now available in the ChatGPT desktop app on Windows. Press both Alt keys to share the frontmost app window with ChatGPT, then ask a question or describe what you want to do. An appshot includes a screenshot and available text from that window. You can customize the shortcut and choose which chat receives it in settings.

Other improvements and bug fixes

• Open files directly from a conversation's Sources panel, or download files that ChatGPT can't preview.
• Assign a Codex Micro key to Insert text and add your own reusable text to the active prompt without sending it.
• Reset a pet to its default size in settings. Custom pets also stay in place when you refresh their artwork.
• Keep unfinished comments on response text when you switch chats and return.
• ChatGPT dictation now respects your saved Main language preference.
• Keep tab widths and scroll positions more stable as you close browser tabs in full and split views.

Install with pip install --upgrade openai-codex==0.154.0 (Python 3.10 or later). This release includes the matching openai-codex-cli-bin==0.154.0 runtime.

Changes

• Add max and ultra reasoning-effort values. (#39662)
• Add ExternalMessage to synchronous and asynchronous run() and turn() calls. External content can start a turn or join an active regular turn with tool-level authority; it does not grant user authorization. Consumers receive independent event streams. (#44086)
• Add include_turns on resume/fork, turn_service_tier for one newly started turn, and source metadata. History selection changes the returned response, not model context. Existing defaults are preserved when these options are omitted. (#44084)
• Refresh generated protocol models and notifications, and preserve completion events that arrive before a turn-start response. (#44032, #44400)

Check these migrations when upgrading:

• HookMetadata wraps its handler in .root. Replace accesses such as hook.command with hook.root.command, checking hook.root.handler_type before reading handler-specific fields.
• Some previously unknown notifications now have typed payloads. Read named fields instead of .params; unknown or invalid payloads still use UnknownNotification.
• Manually constructed or late-joining turn handles receive events from their attachment point. Earlier output is not replayed, so collected results can be partial; attaching after completion can raise TransportClosedError. Use thread.read(include_turns=True) for saved history. Handles returned directly by thread.turn(...) retain events from when their request is sent.

Custom codex_bin overrides require CLI 0.151.0 or newer for ExternalMessage and the new history/per-turn options.

New Features

• Added GPT-6-Astra to the Amazon Bedrock model picker for Mantle and Runtime global/US routes. (#42805)

Bug Fixes

• Corrected GPT-6-Astra’s guidance for asynchronous clarification questions to use the supported tool and recognize that it accepts text only. (#42809)

Changelog

Full Changelog: rust-v0.153.2...rust-v0.153.3

• [0.153 hotfix] Add GPT-6-Astra to Amazon Bedrock catalogs (#42805 @celia-oai)
• [0.153 hotfix] Update Astra async-question guidance (#42809 @rhan-oai)

New Features

• Vim mode now supports undo with u and redo with Ctrl+R, preserving complete drafts including pasted content and attachments. (#41941, #42140)
• The plugin CLI can list, install, and remove plugins from remote marketplaces. (#42150)
• Set tui.auto_recap = false to disable automatic recaps while keeping manual /recap available. (#42101)
• TUI history shows complete patches, input sent to background terminals, and individual completed commands. (#41893, #42107)
• Plus and Team users receive an earlier warning when less than half of their allowance remains in an approximately five-hour usage window. (#42142)

Bug Fixes

• TUI sessions reconnect after an external app-server connection drops, preserving drafts and transcripts while keeping uncertain or queued submissions paused for review. (#41911, #41916, #41918)
• Full Access skips Guardian reviews for confirmation-only actions. User approval mode skips background Guardian scoring and prewarming, while sensitive-action checks and requests for user input retain their existing handling. (#42147, #42256)
• Guardian review history survives compaction, restarts, and user-created forks while respecting rollback boundaries and isolating subagent history. (#41879, #42065)
• Remembered MCP tool approvals are scoped to the selected app account, and relative MCP executable paths start more reliably on macOS. (#42133, #42117)
• Rollout compression includes shared histories, codex exec resume handles compressed rollouts when selecting by working directory, and thread forks work with symlinked session roots. (#42039, #42135)

Configuration and API Updates

• App-server thread metadata includes nullable model and reasoningEffort fields. Structured asynchronous questions are supported through request_user_input_async when enabled by the model catalog. (#42151, #42178)
• tui.disable_paste_burst replaces the top-level setting, which remains supported as a fallback. (#41976)
• Adds the disabled-by-default features.context_management.experimental_mode configuration. When enabled for eligible ChatGPT Plus, Pro, or Pro Lite sessions using the Codex backend, it activates token-budget context, history notes, and the new_context tool. API-key sessions, custom providers, and temporary structured threads remain excluded. (#42385)

Changelog

Full Changelog

• Use shared transcript collection for Guardian reviews (#41870)
• Preserve Guardian review evidence across compaction (#41879)
• Add pinned native voice source preparation (#41884)
• Add native voice dependency build recipe (#41890)
• Retain the MCP client for event streams (#41892)
• Show successful TUI commands individually (#41893)
• Fix Windows native voice dependency builds (#41894)
• Add the voice helper lifecycle foundation (#41897)
• Keep MCP event subscriptions alive after task unloading (#41899)
• Load bounded context after empty wake turns (#41901)
• Add installed voice host lifecycle support (#41902)
• Add a manager for MCP event streams (#41906)
• Avoid scanning archived rollouts when archiving threads (#41908)
• Make permission transforms aware of executor path context (#41909)
• Preserve TUI drafts after app-server disconnects (#41911)
• Persist response token usage in rollout history (#41912)
• Preserve TUI status timing when the status row is hidden (#41915)
• Move the config schema generator into a dedicated crate (#41916)
• Reconnect TUI app-server sessions automatically (#41917)
• Open the agents overview from an empty composer (#41918)
• Restore agent navigation after TUI reconnects (#41923)
• Allow per-call sideband endpoints for existing realtime calls (#41924)
• Record realtime conversation history in Core (#41925)
• Test repository-wide Rust formatter discovery (#41928)
• Use executor path context for permission preapproval (#41929)
• Open the agents overview directly in the reconnect test (#41933)
• Report configured sandbox policy consistently (#41934)
• Omit undersized WAV output from Code Mode (#41936)
• Attach failed Guardian reviews to diagnostic reports (#41937)
• Limit background terminal input previews (#41938)
• Clarify resume guidance in exit summaries (#41940)
• Add Vim undo to the TUI composer (#41941)
• Emit turn cost telemetry for ChatGPT sessions (#41944)
• Expand extension permission regression coverage (#41946)
• Add plugin reconciliation app-server API (#41949)
• Improve tracing for nested tool calls and exec processes (#41950)
• Enforce marketplace source policy for curated plugins (#41953)
• Track TUI starts by app server mode (#41974)
• Move disable_paste_burst under [tui] (#41976)
• Preserve raw response usage metadata (#41980)
• Report turn trigger and source in turn analytics (#42003)
• Share Guardian user-message retention logic (#42031)
• Improve Guardian report diagnostics (#42033)
• Include shared histories in rollout compression (#42039)
• Tag Codex home size metrics with compression state (#42043)
• Add per-account approval settings for apps (#42047)
• Honor explicit account selectors for Apps tool calls (#42054)
• Honor app link settings for MCP tool approvals (#42056)
• Preserve Guardian history across thread reconstruction (#42065)
• Remove selected core test cases (#42066)
• Detect standalone installs from the macOS CLI bundle (#42068)
• Remove redundant test coverage (#42069)
• Detect Vite+-managed Codex installs (#42071)
• Unify Guardian context section collection (#42076)
• Allow models to enable token budgeting by default (#41803)
• Use the async stack budget for approval reviews (#41840)
• Preserve Guardian review evidence across compaction (#41846)
• Preserve Guardian user answers across compaction (#41852)
• Box the session startup future at its API boundary (#41853)
• Preserve Guardian user answers from current history (#41857)
• Preserve user text when Guardian history drops oversized images (#41858)
• Keep history extension tools out of Guardian reviews (#41861)
• Start fresh Vim drafts in Insert mode (#41921)
• Increase Guardian message transcript limits (#41931)

New Features

• Vim mode supports / and ? searches within drafts, highlighted matches, and repeat navigation with n and N. (#41586)
• Rate-limit banners offer actions for checking usage, managing credits, resetting limits, and managing plans. (#41742)
• The terminal UI and codex exec show credential-refresh progress, including Amazon Bedrock reauthentication. (#41239)
• MCP server names can contain :, @, /, and ., supporting package-style names throughout CLI commands and authentication. (#41700)
• Individual MCP tools support an output_token_limit setting, with consistent truncation across session resumes. (#41421)
• App-server clients can configure thread/shellCommand timeouts, including deadlines longer than one hour. (#41384)

Bug Fixes

• Vim-enabled composers now start fresh drafts in Insert mode, including after submitting messages or dispatching slash commands. (#41921)
• Automatic approval reviews can retain longer messages and a larger conversation transcript. (#41931)
• Automatic approval reviews preserve user instructions, answers, and valid authorizations across history compaction. (#41660, #41846, #41852)
• Resumed threads restore their saved working directory when none is supplied, and client metadata updates preserve filesystem permissions. (#41567, #41464)
• MCP tools remain available through cache refreshes and remote plugin changes; authentication retries use refreshed helper-provided headers. (#41336, #41344, #41396, #41400)
• Opening the model picker refreshes available models without losing the highlighted choice. (#41467)
• Fixed Windows sandbox execution with Microsoft Store PowerShell, subprocess hangs on terminal queries, and cursor-related display corruption in older JediTerm terminals. (#41227, #41436, #41673)
• Cloud task requests reject untrusted backend URLs and disable redirects to protect saved credentials. (#41403)

Chores

• The planning tool is disabled by default; enable it with tools.update_plan.enabled = true. (#41744)
• Plugin recommendations begin loading during startup, reducing delays before the first turn. (#41375)

Changelog

Full Changelog: rust-v0.151.0...rust-v0.152.0

New Features

• Added a configurable grace period for discovering tools from optional MCP servers. (#41199)
• Extensions can now inspect or replace MCP tool results before they reach the model. (#41202)
• Plugin catalogs now combine per-repository configuration and report invalid project marketplaces without hiding valid plugins. (#41208)

Bug Fixes

• Preserved restored permission profiles across TUI turns and prevented /cd from weakening sandbox restrictions. (#41192)
• Kept tool availability and reasoning effort correct when switching models or falling back to another model. (#41195, #41206)
• Improved remote sandbox enforcement using the executor’s actual home directory, operating system, and path conventions. (#41196, #41204, #41207, #41209)
• Preserved structured MCP tool and resource errors in app-server responses. (#41196)
• Counted nested subagent token usage toward root goal budgets. (#41183)
• Prevented stale Guardian classifications from authorizing actions after permission state changes. (#41196)

Chores

• Added telemetry for escalated stdin reviews and remote executor MCP discovery. (#41189, #41205)
• Stabilized Guardian WebSocket and core fixture tests under slow or highly concurrent CI. (#41191, #41194)

Changelog

Full Changelog: rust-v0.150.0...rust-v0.151.0

New Features

• Reference other Codex tasks with @ mentions, and ask agents to read, create, or message tasks from the terminal. (#40308, #40315)
• /copy now offers a picker for full responses, individual code blocks, and blockquotes. (#39997)
• Unnamed terminal tasks receive descriptive titles automatically, and /rename suggests an editable title based on the conversation. (#40492, #40495)
• Markdown links appear as clickable labels in supported terminals, with visible URLs retained elsewhere. (#40471)
• Bind shortcuts to cycle permission modes, and use . in Vim mode to repeat your last edit. (#39873, #40521)
• New Interrupt hooks can run commands or MCP handlers when an active top-level turn is interrupted. (#40511)

Bug Fixes

• Untrusted projects no longer supply project-level AGENTS.md instructions, and managed deny-read rules remain enforced after permission changes. (#39837, #40004)
• Improved credential redaction in app-server diagnostics, including provider, authentication refresh, and attestation fields. (#39993)
• Fixed remote MCP bearer-token lookup and required-server startup while preserving compatibility with older executors. (#39926, #39952, #39979)
• Fixed elevated Windows sandbox setup and launch aliases under Unicode user paths. (#39971, #40570)
• Prevented Unix shutdown hangs caused by detached processes retaining a terminal or full terminal output buffers. (#40460)
• Fixed conversation compaction and multi-agent compatibility for Amazon Bedrock models. (#39804, #39825)

Changelog

Full Changelog: rust-v0.149.0...rust-v0.150.0

[followed by detailed changelog entries]

• More browsers: Use the ChatGPT browser extension in Microsoft Edge, Brave, Opera, and Vivaldi, as well as Chrome. Set up your browser in Settings > Computer Use in the ChatGPT desktop app. All five support tab mentions and browser control; Opera doesn’t support side chat.

• Site tools (WebMCP): In the desktop app’s built-in browser, ChatGPT Work and Codex can use tools provided by a website to work with the page. Use GPT-5.6 Sol or GPT-5.6 Terra and update to the latest desktop app. Site tools aren’t available with GPT-5.6 Luna or in Enterprise or Edu workspaces.

• Cloud browser sign-in: On eligible plans, ChatGPT Work on the web, iOS, and Android can ask you to sign in to a supported website through the cloud browser. Enter your details in the sign-in flow, not in the chat. Cloud browser sessions stay separate from your local browser. Website sign-in isn’t available for Enterprise or Edu workspaces.

Availability depends on rollout and workspace settings. Website-access and action-confirmation requirements still apply.

New Features

• Added an interactive codex agents dashboard for searching, starting, opening, renaming, and stopping tasks, with configurable shortcuts. (#39094, #39112, #39114, #39142)
• Added /cd, /pwd, and /cwd commands for managing the working directory in TUI sessions. (#38894)
• Added codex queue for sending messages to existing local or remote sessions. (#39092)
• Expanded Vim editing with character replacement and more change motions such as cw, c$, and cc. (#39661)
• codex doctor now diagnoses endpoint protection, network/proxy failures, desktop app state, and update connectivity. (#38827, #38918, #39060, #39074)
• SDK users can now pass exact CLI config overrides and select max or ultra reasoning effort. (#38817, #39662)

Bug Fixes

• Queued messages now wake idle sessions reliably, resolve duplicate session names more usefully, and preserve pasted or deferred command semantics. (#39034, #39385, #39604)
• Resumed and forked threads now restore their active permission profile instead of silently falling back to current defaults. (#39153)
• Fixed duplicate sub-agent activity and tightened TUI routing for sub-agent notifications and approvals. (#39049, #39088)
• Realtime WebRTC sideband connections now reconnect after unexpected transport loss without dropping pending output. (#39257)
• Inline TUI history now remains available in Windows Terminal scrollback. (#39619)
• Inactive TUI thread replay buffers are now bounded to prevent excessive retained streamed output. (#39081)

Documentation

• Clarified that external contributions should go through issues and design discussion rather than pull requests. (#39089)
• Documented DNS exfiltration risks and trust limitations for secure devcontainers. (#39283)

Changelog

Full Changelog: rust-v0.148.0...rust-v0.149.0

Contributors

copyberry

New features

• Apple Messages: Use the Apple Messages plugin to read and search Messages chats on your Mac and prepare or send messages. It’s available on all plans in the ChatGPT desktop app for macOS. You can use the plugin in ChatGPT Work and Codex. By default, ChatGPT sends messages only after you approve the message and its recipients. See the plugin guide for persistent-approval risks, revocation steps, and the known issue with tasks that disable approval prompts.
• Site co-editing: Where Site collaboration is available, owners can invite active members of the same workspace as editors. Editors can read the Site’s live database data, update the Site, save versions, and publish changes after the owner publishes the Site for the first time. Owners retain control of the audience, settings, analytics, ownership, version restoration, and editor access.
• Editable Site URLs: Where URL editing is available, owners can change an existing Site’s ChatGPT-hosted address without creating another deployment. The previous address redirects to the new URL. Custom domains are a separate, existing feature and aren’t changed by this setting.
• Computer History in Europe: Computer History is now available in the EEA, Switzerland, and the United Kingdom for ChatGPT Pro, Business, and Enterprise users in the ChatGPT desktop app on macOS. It’s off by default and requires Memories. Business and Enterprise administrators must enable access before workspace members can choose to turn it on.
• Shared thread snapshots: On all Codex plans, share a read-only snapshot of a local Codex thread from the ChatGPT desktop app for macOS. The snapshot doesn’t update when the original thread changes. Personal-account links can be opened by anyone with the link; workspace-account links are limited to members of the originating workspace. Codex redacts known secret patterns, but review the shared content because sensitive content may remain. View or revoke links in ChatGPT data controls, under Shared links.
• Unified pinned threads: Keep the same pinned chats across the ChatGPT desktop app and iOS.