RepoJournal
OpenAI

OpenAI

Codex, the SDKs, and the engine behind ChatGPT

Pick a date

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

The Wire · Showcase

CODEX FIXES SUBAGENT DATA MODEL, SHIPS CLOUD CONFIG LAYER STACK

By RepoJournal · Filed · About OpenAI

Codex killed a critical data modeling bug where parent thread relationships got tangled with fork history, then landed the first three PRs of a five-part cloud-managed config overhaul.

The headline fix [1] stops overloading `forked_from_id` to mean two things at once. Guardian and review subagents can now track their parent thread independently from whether they forked the main thread's history. This lands a new `parent_thread_id` field on `SessionMeta` and exposes it through app-server proto, fixing a data model that was breaking subagent relationships downstream.

Meanwhile, the config work ships hard. PR 1 [2] landed the transport layer and bundle endpoint that replaces legacy cloud requirements. PR 2 [3] added the requirements layer composition engine with merge semantics and provenance tracking. PR 3 [4] made enterprise-managed config a first-class layer source with backend-supplied metadata preserved through diagnostics and hook attribution. Together, these three PRs unlock cloud-delivered config as a real system, not a side channel.

On the lighter side, plugin install got faster [5]. Codex stopped forcing directory refresh during auth checks and now uses normal caching instead, cutting unnecessary requests while keeping auth state current. Regression tests lock in the behavior.

Action items

References

  1. [1] store and expose parent_thread_id on Threads ↗ openai/codex
  2. [2] Add config bundle transport types ↗ openai/codex
  3. [3] Compose requirements layers ↗ openai/codex
  4. [4] Add cloud-managed config layer support ↗ openai/codex
  5. [5] [codex] Avoid forced directory refresh during plugin install auth checks (#25381) openai/codex

FAQ

What changed in OpenAI on June 1, 2026?
Codex killed a critical data modeling bug where parent thread relationships got tangled with fork history, then landed the first three PRs of a five-part cloud-managed config overhaul.
What should OpenAI teams do about it?
Review subagent parent_thread_id integration in your session consumers • Watch for PR 4 and 5 of the cloud config stack to ship • Test plugin install performance in your cloud-backed deployments
Which OpenAI repositories shipped on June 1, 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.