Who contributed to OpenAI on May 5, 2026?

3 developers shipped this update, including iceweasel-oai, starr-openai, and fcoury-oai.

What were the notable OpenAI updates?

v6.36.0, v2.34.0, and Fix Windows PTY teardown by preserving ConPTY ownership.

OpenAI

Codex, the SDKs, and the engine behind ChatGPT

Pick a date

Topics: AI / ML Full archive →

The Wire · Showcase

ADMIN API KEYS LAND ACROSS SDKS, CODEX FIXES PTY TEARDOWN ON WINDOWS

By RepoJournal · Filed 06:02 UTC on May 5, 2026 · About OpenAI

3 people shipped this

iceweasel-oai @iceweasel-oai 1 cited

starr-openai @starr-openai 1 cited

fcoury-oai @fcoury-oai 1 cited

Both Node and Python SDKs shipped Admin API Key support [ref:6] [ref:11] while Codex killed a Windows bug that left terminal sessions hanging after exit [ref:2].

The openai-node v6.36.0 [1] and openai-python v2.34.0 [2] releases both introduce Admin API Keys per endpoint, expanding credential management for teams managing multiple organizations. These aren't just version bumps — they're infrastructure changes that unblock multi-tenant deployments. On the Codex side, a critical Windows PTY fix [3] resolves background terminals staying visible after shell exit by preserving ConPTY ownership through teardown, eliminating EOF propagation hangs. The same team landed three prep PRs [4] [5] that restructure process-tool plumbing and thread metadata routing [5], setting up selected environment support without changing behavior yet. A new keymap debug inspector [6] ships in the TUI to help users identify terminal key recognition issues, turning blind bug reports into actionable data. On the infrastructure side, DNS timeout coverage [7] hardens the network proxy against ambient resolver timing issues.

Action items

→ Upgrade openai-node to v6.36.0 and openai-python to v2.34.0 if you manage Admin API keys across endpoints openai/openai-node [plan]
→ If shipping on Windows with Codex, pull the PTY teardown fix [ref:2] before next release openai/codex [plan]
→ Monitor the selected environment prep PRs [ref:3] — follow-up MR will route by environment_id openai/codex [monitor]

References

[1] v6.36.0 ↗ openai/openai-node
[2] v2.34.0 ↗ openai/openai-python
[3] Fix Windows PTY teardown by preserving ConPTY ownership ↗ openai/codex
[4] Prepare selected environment plumbing ↗ openai/codex
[5] codex: route metadata updates through ThreadStore (#20576) openai/codex
[6] feat(tui): add keymap debug inspector ↗ openai/codex
[7] [network-proxy] Cover DNS timeout blocking (#21105) openai/codex

Quick answers

What shipped in OpenAI on May 5, 2026?: Both Node and Python SDKs shipped Admin API Key support [ref:6] [ref:11] while Codex killed a Windows bug that left terminal sessions hanging after exit [ref:2]. In total, 105 commits, 51 pull requests, and 5 releases landed.
Who contributed to OpenAI on May 5, 2026?: 3 developers shipped this update, including iceweasel-oai, starr-openai, and fcoury-oai.
What were the notable OpenAI updates?: v6.36.0, v2.34.0, and Fix Windows PTY teardown by preserving ConPTY ownership.

CODEX HARDENS MCP RUNTIME STABILITY ACROSS FIVE CRITICAL FIXES

OpenAI shipped a coordinated hardening of the Model Context Protocol layer across Codex, patching race conditions in thread startup, resource access, and authority management that could leave stale configurations in production.

ai 66 shipped 1-min read

OpenAI 1 day ago

SDK RELEASES LOCK IN PROMPT CACHING AND SPEND CONTROLS

Both Node and Python clients shipped v6.49.0 and v2.48.0 overnight with admin spend limit APIs and relaxed prompt cache validation.

ai 103 shipped 1-min read

OpenAI 2 days ago

CODEX CONSOLIDATES THREAD STARTUP, NODE SDK ADDS REALTIME AND STANDARD SCHEMA SUPPORT

Codex refactored its threading and database layer for consistency while the Node SDK ships four new schema helpers and fixes audio streaming in a single release push.

ai 120 shipped 1-min read

OpenAI 3 days ago

CODEX FIXES WINDOWS EXECUTION BUGS WHILE TIGHTENING SECURITY REVIEW GATES

Codex shipped critical fixes for Windows process management and terminal navigation that were silently breaking execution sessions on Windows machines.

ai 104 shipped 1-min read

Elsewhere on the wire

AI Agents about 9 hours ago

CLAUDE OPUS 5 LANDS ACROSS THE STACK

The newest Anthropic model is now live in langchain, Cline, and llama-index, with native support for extended reasoning and 1M context windows.

ai-agents 28 shipped 1-min read

Local LLMs about 9 hours ago

OLLAMA LANDS LAGUNA SUPPORT AND CRUSHES MEMORY LEAKS WHILE SGLANG HITS V0.5.16 WITH CONFIDENCE-DRIVEN SPECULATIVE DECODING

Ollama shipped three critical performance and reliability fixes for Metal residency and concurrent access patterns, while SGL-Lang released 0.5.16 with a new speculative algorithm hitting 383.7 tok/s on DeepSeek-V4.