Who contributed to OpenAI on June 9, 2026?

5 developers shipped this update, including celia-oai, rka-oai, jif-oai, won-openai, and viyatb-oai.

What were the notable OpenAI updates?

feat: support oneOf and allOf in tool input schemas, [codex] Enable standalone web search in code mode, and feat: add v2 agent residency lru.

OpenAI

Codex, the SDKs, and the engine behind ChatGPT

Pick a date

Topics: AI / ML Full archive →

The Wire · Showcase

CODEX SHIPS JSON SCHEMA COMPOSITION SUPPORT AND UNLOCKS WEB SEARCH IN CODE MODE

By RepoJournal · Filed 06:02 UTC on June 9, 2026 · About OpenAI

5 people shipped this

celia-oai @celia-oai 1 cited

rka-oai @rka-oai 1 cited

jif-oai @jif-oai 1 cited

won-openai @won-openai 1 cited

viyatb-oai @viyatb-oai 1 cited

Codex now handles the schema shapes that real connectors depend on, while code mode gains direct access to standalone search results.

The oneOf and allOf support landing in Codex [1] closes a gap between what connector golden schemas actually use and what Codex could previously parse. This hits production after the Responses API supports top-level composition, so expect zero disruption and full schema fidelity on the other side. Parallel to that, code mode can now consume plaintext output from standalone web search [2], meaning search results flow directly into nested JavaScript calls instead of getting lost in conversation history. That's a meaningful expansion of what code execution can reach. On the agent side, v2 residency now supports LRU paging [3], letting the stack treat agent identity as separate from whether a thread is currently loaded in ThreadManager. This is the foundation for shifting concurrency counting from "existing agents" to "active execution." Image editing got sharper too: the model now selects exact images for edits instead of Codex inferring them from history [4], which means fewer wrong crops and cleaner intent matching. Finally, sandbox network containment is now properly enforced [5] whenever a configured proxy spins up, ensuring policy-driven network restrictions actually stick across macOS and Linux.

Action items

→ Review and test v2 agent residency paging in your session load patterns openai/codex [plan]
→ Test code mode web search integration before deploying to production openai/codex [plan]
→ Validate sandbox proxy enforcement on your deployment targets openai/codex [monitor]

References

[1] feat: support oneOf and allOf in tool input schemas ↗ openai/codex
[2] [codex] Enable standalone web search in code mode ↗ openai/codex
[3] feat: add v2 agent residency lru ↗ openai/codex
[4] Route image edits through referenced file paths ↗ openai/codex
[5] Enforce configured network proxy in codex sandbox ↗ openai/codex

Quick answers

What shipped in OpenAI on June 9, 2026?: Codex now handles the schema shapes that real connectors depend on, while code mode gains direct access to standalone search results. In total, 38 commits, 38 pull requests, and 5 releases landed.
Who contributed to OpenAI on June 9, 2026?: 5 developers shipped this update, including celia-oai, rka-oai, jif-oai, won-openai, and viyatb-oai.
What were the notable OpenAI updates?: feat: support oneOf and allOf in tool input schemas, [codex] Enable standalone web search in code mode, and feat: add v2 agent residency lru.

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.