Releases · openai/codex

26 Feb 18:12

github-actions

rust-v0.106.0-alpha.11

830a152

0.106.0-alpha.11 Pre-release

Pre-release

Release 0.106.0-alpha.11

Assets 84

25 Feb 17:21

github-actions

rust-v0.105.0

a7eda6a

0.105.0

New Features

The TUI now syntax-highlights fenced code blocks and diffs, adds a /theme picker with live preview, and uses better theme-aware diff colors for light and dark terminals. (#11447, #12581)
You can now dictate prompts by holding the spacebar to record and transcribe voice input directly in the TUI. This feature is still under development; to enable it set features.voice_transcription = true in your config. (#3381)
Multi-agent workflows are easier to run and track: spawn_agents_on_csv can fan out work from a CSV with built-in progress/ETA, and sub-agents are easier to follow with nicknames, a cleaner picker, and visible child-thread approval prompts. (#10935, #12320, #12327, #12332, #12570, #12767)
The TUI picked up new convenience commands: /copy copies the latest complete assistant reply, while /clear and Ctrl-L clear the screen without losing thread context, with /clear also able to start a fresh chat. (#12444, #12520, #12613, #12628)
Approval controls are more flexible: Codex can now ask for extra sandbox permissions for a command, and you can auto-reject specific approval prompt types without turning approvals off entirely. (#11871, #12087)
App-server clients can do more with threads: thread/list can search by title, thread status is exposed in read/list responses and notifications, and thread/resume returns the latest turn inline so reconnects are less lossy. (#11776, #11786, #12578)

Bug Fixes

Long links in the TUI stay clickable when wrapped, which also fixes related clipping and layout issues in several views. (#12067)
Several TUI interaction edge cases were fixed: queued-message editing now works in more terminals, follow-up prompts no longer get stuck if you press Enter while a final answer is still streaming, and approval dialogs now respond with the correct request id. (#12240, #12569, #12746)
@ parsing in the chat composer is more reliable, so commands like npx -y @scope/pkg@latest no longer accidentally open the file picker or block submission. (#12643)
App-server websocket handling is more robust: thread listeners survive disconnects, Ctrl-C waits for in-flight turns before restarting, and websocket clients that send permessage-deflate can connect successfully. (#12373, #12517, #12629)
Linux sandboxed commands now get a minimal /dev, fixing failures in tools that need entropy or other standard device nodes. (#12081)
js_repl now reports uncaught kernel failures more clearly, recovers cleanly afterward, and correctly attaches view_image results from nested tool calls. (#12636, #12725)

Documentation

Added a public security policy with Bugcrowd reporting guidance. (#12193)
Updated install and local workflow docs to use cargo install --locked cargo-nextest and to avoid routine --all-features builds unless you specifically need full feature coverage. (#12377, #12429)

Changelog

Full Changelog: rust-v0.104.0...rust-v0.105.0

#12071 Use V2 websockets if feature enabled @pakrym-oai
#12052 feat(core): zsh exec bridge @owenlin0
#12072 Add message phase to agent message thread item @mousseau-oai
#12077 got rid of experimental_mode for configtoml @won-openai
#12036 codex-api: realtime websocket session.create + typed inbound events @aibrahim-oai
#12073 Add model-visible context layout snapshot tests @charley-oai
#12096 Updated issue labeler script to include safety-check label @etraut-openai
#11944 [js_repl] paths for node module resolution can be specified for js_repl @aaronl-openai
#11798 fix: Restricted Read: /System is too permissive for macOS platform de… @leoshimo-oai
#12015 Enable default status line indicators in TUI config @jif-oai
#12124 feat: phase 1 and phase 2 e2e latencies @jif-oai
#12121 feat: phase 2 usage @jif-oai
#12120 feat: memory usage metrics @jif-oai
#12133 feat: validate agent config file paths @jif-oai
#12137 nit: change model for phase 1 @jif-oai
#12069 Stop filtering model tools in js_repl_tools_only mode @fjord-oai
#12135 feat: better slug for rollout summaries @jif-oai
#12157 Disable collab tools during review delegation @jif-oai
#11802 Fixed a hole in token refresh logic for app server @etraut-openai
#12105 fix: file watcher @jif-oai
#12167 memories: bump rollout summary slug cap to 60 @zuxin-oai
#12177 js_repl: canonicalize paths for node_modules boundary checks @fjord-oai
#12025 app-server support for Windows sandbox setup. @iceweasel-oai
#12187 fix: Remove citation @zuxin-oai
#12180 [apps] Temporary app block. @mzeng-openai
#11786 app-server: expose loaded thread status via read/list and notifications @euroelessar
#12038 state: enforce 10 MiB log caps for thread and threadless process logs @charley-oai
#12203 Stabilize app-server detached review and running-resume tests @charley-oai
#12211 [apps] Update apps allowlist. @mzeng-openai
#12081 fix(linux-sandbox): mount /dev in bwrap sandbox @viyatb-oai
#12164 Update docs links for feature flag notice @etraut-openai
#12231 Adjust memories rollout defaults @jif-oai
#12152 feat: sub-agent injection @jif-oai
#12233 feat: add shell snapshot failure reason @jif-oai
#12193 docs: add codex security policy @viyatb-oai
#12228 feat: add configurable write_stdin timeout @jif-oai
#11787 Adjust MCP tool approval handling for custom servers @colby-oai
#12179 Move previous turn context tracking into ContextManager history @charley-oai
#12244 Restore phase when loading from history @mousseau-oai
#12101 client side modelinfo overrides @sayan-oai
#12251 Add configurable agent spawn depth @jif-oai
#12254 chore: increase stack size for everyone @jif-oai
#12250 feat: no timeout mode on ue @jif-oai
#12255 chore: consolidate new() and initialize() for McpConnectionManager @bolinfest
#12256 Revert "feat: no timeout mode on ue" @jif-oai
#12258 Undo stack size Bazel test hack @charley-oai
#12264 Fix flaky test @jif-oai
#12087 feat: add Reject approval policy with granular prompt rejection controls @bolinfest
#12253 Skip removed features during metrics emission @jif-oai
#12265 Clarify cumulative proposed_plan behavior in Plan mode @charley-oai
#12266 app-server tests: reduce intermittent nextest LEAK via graceful child shutdown @bolinfest
#11953 ws turn metadata via client_metadata @pash-openai
#11778 fix(bazel): replace askama templates with include_str! in memories @rupurt
#11382 Add configurable MCP OAuth callback URL for MCP login @dkumar-oai
#12274 [bazel] Fix proc_macro_dep libs @zbarsky-openai
#12054 feat(config): add permissions.network proxy config wiring @viyatb-oai
#12269 tests(thread_resume): interrupt running turns in resume error-path tests @bolinfest
#12009 Update pnpm versions to fix cve-2026-24842 @mjr-openai
#12080 skill-creator: lazy-load PyYAML in frontmatter parsing @xl-openai
#12271 tests: centralize in-flight turn cleanup helper @bolinfest
#12286 app-server: fix flaky list_apps_returns_connectors_with_accessible_flags test @bolinfest
#11776 app-server: improve thread resume rejoin flow @maxj-oai
#11822 [apps] Store apps tool cache in disk to reduce startup time. @mzeng-openai
#12221 memories: add rollout_summary_file header to raw memories and tune prompts @zuxin-oai
#12309 Set memories phase reasoning effort constants @jif-oai
#12315 chore: nit explorer @jif-oai
#12320 feat: add nick name to sub-agents @jif-oai
#12328 chore: better agent names @jif-oai
#12326 disable collab for phase 2 @jif-oai
#12267 Add MCP server context to otel tool_result logs @colby-oai
#12327 feat: cleaner TUI for sub-agents @jif-oai
#12332 feat: better agent picker in TUI @jif-oai
#12344 feat: do not enqueue phase 2 if not necessary @jif-oai
#12340 fix: simplify macOS sleep inhibitor FFI @yvolovich-cyber
#12347 fix: nick name at thread/read @jif-oai
#12294 Reuse connection between turns @pakrym-oai
#12287 app-server: add JSON tracing logs @maxj-oai
#12140 Refactor network approvals to host/protocol/port scope @viyatb-oai
#12300 js_repl: block wrapped payload prefixes in grammar @fjord-oai
#11368 fix(network-proxy): add unix socket allow-all and update seatbelt rules @viyatb-oai
#12275 js_repl: remove codex.state helper references @fjord-oai
#12289 CODEX-4927: Surface local login entitlement denials in browser @daniel-oai
#12205 ci(bazel): install Node from node-version.txt in remote image @fjord-oai
#12312 feat: add config allow_login_shell @jif-oai
#12086 [apps] Implement apps configs. @mzeng-openai
#12353 fix(core): require approval for destructive MCP tool calls @colby-oai
#12218 app-server: harden disconnect cleanup paths @maxj-oai
#12291 core tests: use hermetic mock server in review suite @viyatb-oai
#12374 [apps] Enforce simple logo url format. @mzeng-openai
#12377 docs: use --locked when installing cargo-nextest @derekf-oai
#12370 Add ability to attach extra files to feedback @pakrym-oai
#12306 Move sanitizer into codex-secrets @viyatb-oai
#12379 clarify model_catalog_json only applied on startup @sayan-oai
#12307 Show model/reasoning hint when switching modes @charley-oai
#12403 [apps] Fix gateway url. @mzeng-openai
#12240 fix(tui): queued-message edit shortcut unreachable in some terminals @fcoury
#12405 [apps] Bump MCP tool call timeout. @mzeng-openai
#12406 fix: explicitly list name collisions in JSON schema generation @bolinfest
#12301 Add field to Thread object for the latest rename set for a given thread @natea-oai
#12268 Wire realtime api to core @aibrahim-oai
#12415 fix(nix): include libcap dependency on linux builds @rgodha24
#12416 Add experimental realtime websocket URL override @aibrahim-oai
#12303 Improve Plan mode reasoning selection flow @charley-oai
#12418 Add experimental realt...