Who contributed to OpenAI on May 6, 2026?

2 developers shipped this update, including fcoury-oai and rhan-oai.

What were the notable OpenAI updates?

Enable V8 sandboxing for source-built builds (#21146), fix(sandboxing): Bound advisory system bwrap startup probe (#20111), and fix(linux-sandbox): avoid panic on bwrap build failures (#21127).

OpenAI

Codex, the SDKs, and the engine behind ChatGPT

Pick a date

Topics: AI / ML Full archive →

The Wire · Showcase

V8 SANDBOXING GOES LIVE IN SOURCE BUILDS

By RepoJournal · Filed 06:03 UTC on May 6, 2026 · About OpenAI

2 people shipped this

fcoury-oai @fcoury-oai 1 cited

rhan-oai @rhan-oai 1 cited

Codex ships the first phase of in-process V8 sandboxing, hardening the JavaScript execution layer while two critical startup bugs get fixed.

The V8 sandboxing rollout [1] lands build system and Rust feature plumbing to run V8 in a hardened sandbox for source-built Bazel paths, deliberately keeping published rusty_v8 artifacts on their current non-sandboxed contract so this ships independently before changing any released artifacts. Two blocking sandbox issues got resolved overnight: a startup probe on NFS/autofs hosts that was stalling Codex launch [2] now times out gracefully instead of blocking initialization, and bubblewrap build failures [3] no longer panic but propagate as ordinary sandbox failures with proper symlink carveout handling. On the app-server side, /diff commands now route through the workspace abstraction [4] so remote sessions compute diffs against the correct machine, and Turn payloads added an itemsView enum [5] to disambiguate unloaded, summary, and fully hydrated turn states — unblocking future lazy-loading work. This is a coordinated hardening sprint across the execution sandbox, startup paths, and remote workspace semantics.

Action items

→ Verify V8 sandboxing builds in your local Bazel workflow and flag any feature gate issues before next release candidate openai/codex [plan]
→ Test /diff behavior on remote app-server sessions to confirm Git operations run against active workspace openai/codex [monitor]
→ Review Turn.items consumers to handle new itemsView metadata in app-server payloads openai/codex [plan]

References

[1] Enable V8 sandboxing for source-built builds (#21146) openai/codex
[2] fix(sandboxing): Bound advisory system bwrap startup probe (#20111) openai/codex
[3] fix(linux-sandbox): avoid panic on bwrap build failures (#21127) openai/codex
[4] feat(tui): route /diff through workspace commands ↗ openai/codex
[5] add turn items view to app-server turns ↗ openai/codex

Quick answers

What shipped in OpenAI on May 6, 2026?: Codex ships the first phase of in-process V8 sandboxing, hardening the JavaScript execution layer while two critical startup bugs get fixed. In total, 50 commits, 50 pull requests, and 3 releases landed.
Who contributed to OpenAI on May 6, 2026?: 2 developers shipped this update, including fcoury-oai and rhan-oai.
What were the notable OpenAI updates?: Enable V8 sandboxing for source-built builds (#21146), fix(sandboxing): Bound advisory system bwrap startup probe (#20111), and fix(linux-sandbox): avoid panic on bwrap build failures (#21127).

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 10 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 10 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.