AI Changelog
Back to feed

0.52.0

Oct 30, 2025
OpenAI/Codex CLICLIvrust-v0.52.0
View original

Highlights

  • TUI polish: queued messages visible during streaming; Windows auto-mode guidance; undo op; compaction prompt configurable.
  • Images: client-side image resizing; prevent crashes with MIME verification; SDK image forwarding tests re-enabled.
  • Execute commands directly with !<cmd>
  • Ability to buy credits for more usage. More information

Other improvements

  • Auth: new CLI auth storage abstraction with keyring support and configurable backend.
  • App server: GetConversationSummary RPC; fetch/resume conversation summaries by id; refined account/protocol types.
  • Execution/events: item streaming events; ghost commits; clearer deprecation warnings; better token refresh handling and logs.
  • Conversation management: centralized truncation for all tool outputs; filter out prior-turn reasoning; annotate conversations with model_provider.

List of merged PRs

  1. Log HTTP Version in https://github.com/openai/codex/pull/5475
  2. feat: local tokenizer in https://github.com/openai/codex/pull/5508
  3. chore: drop approve all in https://github.com/openai/codex/pull/5503
  4. [MCP] Add support for specifying scopes for MCP oauth in https://github.com/openai/codex/pull/5487
  5. chore: clean spec tests in https://github.com/openai/codex/pull/5517
  6. [MCP] Remove the legacy stdio client in favor of rmcp in https://github.com/openai/codex/pull/5529
  7. [app-server] send account/rateLimits/updated notifications in https://github.com/openai/codex/pull/5477
  8. Fix unified exec session id test helper in https://github.com/openai/codex/pull/5535
  9. [app-server] add new account method API stubs in https://github.com/openai/codex/pull/5527
  10. chore: clean handle_container_exec_with_params in https://github.com/openai/codex/pull/5516
  11. fix: approval issue in https://github.com/openai/codex/pull/5525
  12. tui: show aggregated output in display in https://github.com/openai/codex/pull/5539
  13. Handle cancelling/aborting while processing a turn in https://github.com/openai/codex/pull/5543
  14. chore: testing on apply_path in https://github.com/openai/codex/pull/5557
  15. feat: use actual tokenizer for unified_exec truncation in https://github.com/openai/codex/pull/5514
  16. feat: end events on unified exec in https://github.com/openai/codex/pull/5551
  17. Use Option symbol for mac key hints in https://github.com/openai/codex/pull/5582
  18. Moving token_info to ConversationHistory in https://github.com/openai/codex/pull/5581
  19. [MCP] Update rmcp to 0.8.3 in https://github.com/openai/codex/pull/5542
  20. [MCP] Improve startup errors for timeouts and github in https://github.com/openai/codex/pull/5595
  21. fix: flaky tests in https://github.com/openai/codex/pull/5625
  22. adding messaging for stale rate limits + when no rate limits are cached in https://github.com/openai/codex/pull/5570
  23. Add CodexHttpClient wrapper with request logging in https://github.com/openai/codex/pull/5564
  24. [app-server] fix account/read response annotation in https://github.com/openai/codex/pull/5642
  25. Add instruction for upgrading codex with brew in https://github.com/openai/codex/pull/5640
  26. Log more types of request IDs in https://github.com/openai/codex/pull/5645
  27. Added model summary and risk assessment for commands that violate sandbox policy in https://github.com/openai/codex/pull/5536
  28. [MCP] Redact environment variable values in /mcp and mcp get in https://github.com/openai/codex/pull/5648
  29. [MCP] Properly gate login after mcp add with experimental_use_rmcp_client in https://github.com/openai/codex/pull/5653
  30. [codex][app-server] introduce codex/event/raw_item events in https://github.com/openai/codex/pull/5578
  31. Fixed flaky unit test in https://github.com/openai/codex/pull/5654
  32. Improve feedback in https://github.com/openai/codex/pull/5661
  33. Followup feedback in https://github.com/openai/codex/pull/5663
  34. Skip flaky test in https://github.com/openai/codex/pull/5680
  35. Added support for sandbox_mode in profiles in https://github.com/openai/codex/pull/5686
  36. Reduced runtime of unit test that was taking multiple minutes in https://github.com/openai/codex/pull/5688
  37. [MCP] Minor docs clarifications around stdio tokens in https://github.com/openai/codex/pull/5676
  38. fix: use codex-exp prefix for experimental models and consider codex- models to be production in https://github.com/openai/codex/pull/5797
  39. fix: revert "[app-server] fix account/read response annotation (#5642)" in https://github.com/openai/codex/pull/5796
  40. Add feedback upload request handling in https://github.com/openai/codex/pull/5682
  41. feat: annotate conversations with model_provider for filtering in https://github.com/openai/codex/pull/5658
  42. feat: update NewConversationParams to take an optional model_provider in https://github.com/openai/codex/pull/5793
  43. feat: undo wiring in https://github.com/openai/codex/pull/5630
  44. feat: async ghost commit in https://github.com/openai/codex/pull/5618
  45. feat: TUI undo op in https://github.com/openai/codex/pull/5629
  46. nit: doc on session task in https://github.com/openai/codex/pull/5809
  47. chore: undo nits in https://github.com/openai/codex/pull/5631
  48. fix: test yield time in https://github.com/openai/codex/pull/5811
  49. feat: introduce GetConversationSummary RPC in https://github.com/openai/codex/pull/5803
  50. feat: return an error if unknown enabled/disabled feature in https://github.com/openai/codex/pull/5817
  51. Made token refresh code resilient to missing id_token in https://github.com/openai/codex/pull/5782
  52. [Auth] Introduce New Auth Storage Abstraction for Codex CLI in https://github.com/openai/codex/pull/5569
  53. feat: support verbosity in model_family in https://github.com/openai/codex/pull/5821
  54. [Auth] Add keyring support for Codex CLI in https://github.com/openai/codex/pull/5591
  55. Centralize truncation in conversation history in https://github.com/openai/codex/pull/5652
  56. fix: move account struct to app-server-protocol and use camelCase in https://github.com/openai/codex/pull/5829
  57. [MCP] Render MCP tool call result images to the model in https://github.com/openai/codex/pull/5600
  58. fix image drag drop in https://github.com/openai/codex/pull/5794
  59. Truncate the content-item for mcp tools in https://github.com/openai/codex/pull/5835
  60. feat(tui): clarify Windows auto mode requirements in https://github.com/openai/codex/pull/5568
  61. [Auth] Choose which auth storage to use based on config in https://github.com/openai/codex/pull/5792
  62. chore: speed-up pipeline in https://github.com/openai/codex/pull/5812
  63. chore: drop useless shell stuff in https://github.com/openai/codex/pull/5848
  64. chore: use anyhow::Result for all app-server integration tests in https://github.com/openai/codex/pull/5836
  65. chore: decompose submission loop in https://github.com/openai/codex/pull/5854
  66. fix advanced.md in https://github.com/openai/codex/pull/5833
  67. tui: show queued messages during response stream in https://github.com/openai/codex/pull/5540
  68. Filter out reasoning items from previous turns in https://github.com/openai/codex/pull/5857
  69. revert #5812 release file in https://github.com/openai/codex/pull/5887
  70. remove beta experimental header in https://github.com/openai/codex/pull/5892
  71. Fixed bug that results in a sporadic hang when attaching images in https://github.com/openai/codex/pull/5891
  72. verify mime type of images in https://github.com/openai/codex/pull/5888
  73. [app-server] Annotate more exported types with a title in https://github.com/openai/codex/pull/5879
  74. [App Server] Allow fetching or resuming a conversation summary from the conversation id in https://github.com/openai/codex/pull/5890
  75. [codex][app-server] resume conversation from history in https://github.com/openai/codex/pull/5893
  76. Refresh tokens more often and log a better message when both auth and token refresh fails in https://github.com/openai/codex/pull/5655
  77. fix(windows-path): preserve PATH order; include core env vars in https://github.com/openai/codex/pull/5579
  78. chore: merge git crates in https://github.com/openai/codex/pull/5909
  79. feat: deprecation warning in https://github.com/openai/codex/pull/5825
  80. feat: add output even in sandbox denied in https://github.com/openai/codex/pull/5908
  81. Add missing "nullable" macro to protocol structs that contain optional fields in https://github.com/openai/codex/pull/5901
  82. Add a wrapper around raw response items in https://github.com/openai/codex/pull/5923
  83. fix: icu_decimal version in https://github.com/openai/codex/pull/5919
  84. chore: config editor in https://github.com/openai/codex/pull/5878
  85. Delegate review to codex instance in https://github.com/openai/codex/pull/5572
  86. [exec] Add MCP tool arguments and results in https://github.com/openai/codex/pull/5899
  87. Add item streaming events in https://github.com/openai/codex/pull/5546
  88. Re-enable SDK image forwarding test in https://github.com/openai/codex/pull/5934
  89. ignore agent message deltas for the review mode in https://github.com/openai/codex/pull/5937
  90. asdf in https://github.com/openai/codex/pull/5940
  91. Add debug-only slash command for rollout path in https://github.com/openai/codex/pull/5943
  92. Add debug-only slash command for rollout path in https://github.com/openai/codex/pull/5936
  93. Send delegate header in https://github.com/openai/codex/pull/5942
  94. chore: unify config crates in https://github.com/openai/codex/pull/5958
  95. Pass initial history as an optional to codex delegate in https://github.com/openai/codex/pull/5950
  96. feat: compaction prompt configurable in https://github.com/openai/codex/pull/5959
  97. nit: log rmcp_client in https://github.com/openai/codex/pull/5978
  98. chore: testing on freeform apply_patch in https://github.com/openai/codex/pull/5952
  99. [codex] add developer instructions in https://github.com/openai/codex/pull/5897
  100. [app-server] remove serde(skip_serializing_if = "Option::is_none") annotations in https://github.com/openai/codex/pull/5939
← Back to feed