RepoJournal
OpenAI

OpenAI

Codex, the SDKs, and the engine behind ChatGPT

Keep up with OpenAI in about 3 minutes a day: what actually shipped — the commits, pull requests, releases, and security advisories that matter.

One email a day. Unsubscribe in one click.

Pick a date

  1. Fri 10 jul
  2. Sat 11 jul
  3. Sun 12 jul
  4. Mon 13 jul
  5. Tue 14 jul
  6. Wed 15 jul
  7. Thu 16 jul
  8. Sat 18 jul
  9. Sun 19 jul
  10. Mon 20 jul
  11. Tue 21 jul
  12. Wed 22 jul
  13. Thu 23 jul
  14. Fri 24 jul
  15. Sat 25 jul
  16. Sun 26 jul
  17. Mon 27 jul
  18. Tue 28 jul
  19. Wed 29 jul
  20. Thu 30 jul
  21. Fri 31 jul
  22. Sat 1 aug
  23. Sun 2 aug
  24. Mon 3 aug
  25. Tue 4 aug
  26. Wed 5 aug
  27. Thu 6 aug
  28. Fri 7 aug
  29. Sat 8 aug
  30. Today 9 aug
Topics: AI / ML Full archive →

The Wire · Showcase

CODEX HARDENS STREAMS AND SANDBOXING AS SDK CLEANUP ACCELERATES

By RepoJournal · Filed · About OpenAI

Hayden copyberry felixxia-oai sayan-oai
4 people shipped this
Hayden @HAYDEN-OAI 5 cited copyberry @copyberry 3 cited felixxia-oai @felixxia-oai 1 cited sayan-oai @sayan-oai 1 cited

Codex shipped resilience for network failures and executor-native sandboxing while the Node SDK continues its oxlint cleanup sprint across five new lint rule enablements.

Codex is now keeping response streams alive through connection failures [1], retrying with exponential delays up to 60 seconds and showing users a reconnecting message instead of dropping the connection entirely. That same resilience push includes delegation of remote process sandboxing to the executor [2], preserving executor-native working directories and sandbox intent rather than resolving them through the host platform. On the code execution side, interrupted turns can now terminate active code-mode cells [3] when the feature is enabled, keeping the session alive for later use. The architecture work continues with a loader for executor-local config layers [4] that returns raw configuration with sources and precedence order, and a consolidation of prompt injection logic into the skills extension [5]. Over in openai-node, the SDK is grinding through its oxlint cleanup stack: five separate PRs landed this period enabling jsdoc/require-yields-description [6], no-sequences [7], unicorn/no-hex-escape [9], and unicorn/prefer-includes [10], while a broader reorganization moved root-level examples into API-specific folders for assistants, audio, chat-completions, and other areas [8]. Across both repos, 90 commits and 94 PRs landed over the last 24 hours with 3 releases.

One email a day. Unsubscribe in one click.

Action items

References

  1. [1] Keep response streams alive through connection failures ↗ openai/codex
  2. [2] Delegate remote process sandboxing to the executor ↗ openai/codex
  3. [3] Interrupt active code-mode cells with their turn ↗ openai/codex
  4. [4] Add a loader for executor-local config layers (#37406) openai/codex
  5. [5] Move host skill prompt injection into the skills extension (#37503) openai/codex
  6. [6] chore(lint): enable jsdoc/require-yields-description (#2100) openai/openai-node
  7. [7] chore(lint): enable no-sequences (#2123) openai/openai-node
  8. [8] [SDK-196] Organize examples by API area (#2127) openai/openai-node
  9. [9] chore(lint): enable unicorn/no-hex-escape (#2129) openai/openai-node
  10. [10] chore(lint): enable unicorn/prefer-includes (#2130) openai/openai-node

Quick answers

What shipped in OpenAI on August 8, 2026?
Codex shipped resilience for network failures and executor-native sandboxing while the Node SDK continues its oxlint cleanup sprint across five new lint rule enablements. In total, 90 commits, 94 pull requests, and 3 releases landed.
Who contributed to OpenAI on August 8, 2026?
4 developers shipped this update, including felixxia-oai, sayan-oai, copyberry, and Hayden.
What were the notable OpenAI updates?
Keep response streams alive through connection failures, Delegate remote process sandboxing to the executor, and Interrupt active code-mode cells with their turn.

More from OpenAI

OpenAI OpenAI

CODEX HARDENS IDENTITY AND AUDIT LAYER ACROSS PLUGIN, HOOK, AND PROCESS EXECUTION

Codex shipped five infrastructure changes overnight that lock down environment isolation, plugin failure tracking, and workload identity token handling across the entire execution stack.

copyberry
ai 16 shipped
OpenAI OpenAI

CODEX SHIPS ROLLOUT MIGRATION FRAMEWORK AND CONTEXT WINDOW TRACKING

Codex rolled out a complete migration toolkit for legacy session handling while fixing a critical flaw in how forked subagents track their context windows.

copyberry TAFOYA-OAI apcha-oai
ai 74 shipped
OpenAI OpenAI

CODEX HARDENS SKILL SELECTION AND SESSION LOOKUP, NODE SDK CLOSES LINTER GAPS

Codex decouples explicit skill selection into a dedicated crate while tightening TUI session validation, as the Node SDK completes its oxlint cleanup stack across five rule categories.

felixxia-oai github-actions[bot] copyberry Hayden
ai 118 shipped
OpenAI OpenAI

CODEX HARDENS SECURITY, SDKS MODERNIZE TOOLCHAINS

Codex now prompts users before trusting local directories, closing a prompt injection exposure, while both Node and Python SDKs shift to Oxlint and Oxfmt tooling.

efrazer-oai Eric Traut copyberry HAYDEN-OAI +1
ai 114 shipped

Daily updates, in your inbox

Follow OpenAI

Keep up with OpenAI in about 3 minutes a day: what actually shipped — the commits, pull requests, releases, and security advisories that matter.

or

One email a day. Unsubscribe in one click. Read a past issue →

Elsewhere on the wire

AI Agents AI Agents

CLINE FIXES ABORT-WINDOW PROMPT LOSS, CLIPBOARD IMAGES LAND ON DESKTOP

Cline's core agent now preserves queued prompts when you hit stop, and desktop users can finally paste screenshots directly into the composer.

dependabot malhotra5 Saoud Rizwan
ai-agents 40 shipped
Local LLMs Local LLMs

OLLAMA CUTS MODEL CONFIG BLOAT 100KB+ PER MODEL, LLAMA.CPP SHIPS TOOL ISOLATION AND CUDA FUSION GAINS

Ollama eliminated redundant manifest data ballooning config blobs on safetensors models, while llama.cpp landed docker-backed tool isolation and fused three GPU kernels into one.

Daniel Hiltgen github-actions[bot] grafail Isotr0py +7
llms 102 shipped
CachyOS @CachyOS

WIKI TRANSLATION PUSH HITS SPANISH AND PORTUGUESE, NVIDIA DRIVERS GET 7.2 SUPPORT

CachyOS documentation expands globally with AI-assisted Spanish and Portuguese updates while the core team hardens nvidia-470xx for kernel 7.2.

witchlliee github-actions Vasiliy Stelmachenok Vladislav Nepogodin +3
infra 73 shipped
Elixir & Phoenix Elixir & Phoenix

MACRO.TO_STRING BUG BREAKS FOR COMPREHENSIONS WITH TRAILING KEYWORDS

Elixir's Macro.to_string/1 silently corrupts for-comprehensions that use do-end blocks with non-block trailing keywords, rendering valid code that no longer compiles.

makoto-developer
elixir 2 shipped
Want every project, not just this one?