RepoJournal
OpenAI

OpenAI

Codex, the SDKs, and the engine behind ChatGPT

Pick a date

  1. Wed 20 may
  2. Thu 21 may
  3. Sat 23 may
  4. Mon 25 may
  5. Tue 26 may
  6. Wed 27 may
  7. Thu 28 may
  8. Fri 29 may
  9. Sat 30 may
  10. Sun 31 may
  11. Mon 1 jun
  12. Mon 8 jun
  13. Tue 9 jun
  14. Wed 10 jun
  15. Thu 11 jun
  16. Fri 12 jun
  17. Sat 13 jun
  18. Sun 14 jun
  19. Mon 15 jun
  20. Tue 16 jun
  21. Wed 17 jun
  22. Thu 18 jun
  23. Fri 19 jun
  24. Sat 27 jun
  25. Sun 28 jun
  26. Mon 29 jun
  27. Tue 30 jun
  28. Wed 1 jul
  29. Thu 2 jul
  30. Today 3 jul

The Wire · Showcase

CODEX SHIPS SAFETY LINK FIX AND MAX REASONING ALIGNMENT

By RepoJournal · Filed · About OpenAI

Safety check surfaces in Codex were routing users to stale documentation, and the reasoning picker was mishandling the max effort level—both are now fixed.

The most critical ship is the safety check link update [1], which patches broken Trusted Access URLs and ensures the Help Center surfaces consistently across Bio and Cyber safety surfaces in the TUI. This closes a gap in the previous #30317 work where retry and wait behavior are preserved. In parallel, Codex now treats max as a first-class reasoning effort [2], fixing a catalog alignment problem where Bedrock GPT-5.6 advertises max but Codex was rendering it as lowercase while treating it as an opaque custom value. The fix adds typed ReasoningEffort::Max parsing and serialization without changing the wire protocol or persisted representation. On the infrastructure side, Nagle has been disabled unconditionally on Rendezvous WebSockets [3] to improve latency on exec-server connections, with TCP_NODELAY enabled at both connection endpoints and no feature flags or telemetry overhead added.

Action items

References

  1. [1] Update safety check links ↗ openai/codex
  2. [2] [codex] Treat max as a first-class reasoning effort ↗ openai/codex
  3. [3] [codex] disable Nagle on Rendezvous WebSockets (#30269) openai/codex

FAQ

What changed in OpenAI on June 30, 2026?
Safety check surfaces in Codex were routing users to stale documentation, and the reasoning picker was mishandling the max effort level—both are now fixed.
What should OpenAI teams do about it?
Review and test the safety link updates [ref:1] in your next staging cycle • Verify reasoning effort picker now displays max with productized label [ref:2] • Monitor WebSocket latency after Nagle disable lands in production [ref:3]
Which OpenAI repositories shipped on June 30, 2026?
openai/codex

Related across the cluster

For your repos

The showcase is a teaser.
Your wire is the product.

Same engine. Different stack. Below: what changes when the wire is yours.

Showcase wire

  • •14 famous open source orgs
  • •One wire per day
  • •Public, generic
  • •Read on the web, when you remember

Your wire

  • →Up to 1,500 of your repos - orgs, deps, vendors
  • →Morning and evening briefs
  • →Action items routed to your team
  • →Slack delivery, email, breaking-news CVE alerts

Want a hands-on demo first? Ask a current user for an invite link.