RepoJournal
OpenAI OpenAI
133 wires and counting

$ follow OpenAI

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

or

fair warning: these emails are deeply technical. diffs, version numbers, CVEs, benchmark deltas. if that's not your idea of a good read, this isn't your newsletter.

Folds into your digest — weekly by default, monthly if you prefer. Unsubscribe in one click.

$ status

wire 2026-07-08
stories 66
RSS Archive Reviews Topics

© 2026 RepoJournal Home Showcase Explore How it works Privacy

$ the-wire · showcase

INTEL MACOS CODEX CRASHES ON V8 STARTUP, SIGNING FIX SHIPS

By RepoJournal · Filed · About OpenAI · Composed from the cited sources · methodology

owenlin0 celia-oai rafael-jac
3 people shipped this
owenlin0 @owenlin0 2 cited celia-oai @celia-oai 1 cited rafael-jac @rafael-jac 1 cited

Code Mode tool calls are crashing on Intel macOS because V8's code range allocation triggers Hardened Runtime rejection without the right entitlement.

OpenAI shipped an emergency fix [1] that adds the missing Intel V8 signing entitlement to Codex release binaries. The crash happens during the first tool call when V8 tries to make a non-JIT memory reservation executable, and the current signature only contains the JIT-specific entitlement. This is an immediate production issue for anyone running Codex on Intel Macs with the latest GPT-5.6 sol tool calls.

In parallel, the core team is standardizing how collaboration and sub-agent activity items flow through the system. Both the canonical collab tool call items [2] and the canonical sub-agent activity items [3] are now emitting through the proper TurnItem lifecycle instead of legacy events. App-server v2 consumes these canonical items directly, applies thread-watch cleanup from completed items, and ignores the mapped legacy events. This architectural consolidation reduces event complexity and tightens the contract between core and app-server.

A refactor extracted remote compaction request attempts [4] as a behavior-preserving change ahead of follow-up work on compaction fallbacks. The login system now supports hosted success redirects [5], letting app-server callers opt into a hosted page instead of localhost, while persisting credentials before the redirect.

$ ls openai/reviews/ # the step-back reads

the week in review Codex 0.157.0 ships GPT-6 Sol and Luna the month in review Agents API lands in both OpenAI SDKs
Topics: AI / ML Full archive → Weekly reviews →

Quick answers

What shipped in OpenAI on July 8, 2026?
Code Mode tool calls are crashing on Intel macOS because V8's code range allocation triggers Hardened Runtime rejection without the right entitlement. In total, 32 commits, 32 pull requests, and 2 releases landed.
Who contributed to OpenAI on July 8, 2026?
3 developers shipped this update, including owenlin0, celia-oai, and rafael-jac.
What were the notable OpenAI updates?
fix(release): add missing Intel V8 signing entitlement (#30953), feat(core): emit canonical collab tool call items, and feat(core): emit canonical sub-agent activity items.
♥

$ log OpenAI --recent

Computer use lands in both SDKs, WebSocket lane IDs stay reserved 152 shipped Live and Realtime queues stop replaying uncertain sends 131 shipped Base URL query bugs fixed across Node and Python realtime clients 86 shipped Codex keeps WebSocket turns alive on steer, fixes Windows CLI launch 60 shipped

$ log --all # elsewhere on the wire

@openclaw openclaw 2026.9.7 hardens update rollback, and Peekaboo ships GPT-6.1 Sol Agentic Coding Enterprise permission defaults get teeth as Codex, Gemini CLI and Goose ship fixes

$ ls showcases/

Rails Django Laravel Node.js Go Vercel Vue.js Kubernetes HashiCorp Arch Linux Linux Anthropic

$ init your-desk

RepoJournal can write one of these for any repo you care about. Sign in with GitHub, pick your repos and sources, and your desk gets a wire like this one every morning.