RepoJournal
OpenAI OpenAI
150 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 2026-05-16
stories 56
RSS Archive Reviews

© 2026 RepoJournal Home Showcase How it works Privacy

$ the-wire · showcase

CODEX COMPLETES PERMISSIONS MIGRATION, NODE AND PYTHON SDKS SHIP SERVICE TIER SUPPORT

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

stainless-app[bot]
1 person shipped this
stainless-app[bot] @stainless-app[bot] 2 cited

OpenAI shipped matching SDK updates across Node and Python while Codex finished moving its permissions model into production, cleaning up years of legacy sandbox code in the process.

The Codex team landed five connected commits that finish migrating the permissions system away from the old `SandboxPolicy` model [1] [2] [3] [4] [5]. Memory prompt injection now routes through the app-server extension registry instead of living inline in codex-core [1], permission profiles no longer leak from the thread lifecycle API [3], telemetry tags derive directly from the active profile [4], and the legacy `PermissionsInstructions` conversion helper got removed entirely [5]. This is cleanup work that matters: it means the codebase no longer has two competing mental models for how permissions work. On the SDK side, both openai-node v6.38.0 [6] [7] and openai-python v2.37.0 [8] [9] shipped the same day with a new `service_tier` parameter on the responses compact method. The Python release also added eager validation for pydantic iterators, which will catch type issues before they hit production. These are synchronized releases, so Node and Python users stay aligned on API coverage.

Action items

References

  1. [1] Move memory prompt injection to app-server extension (#22841) ↗ openai/codex
  2. [2] core: construct test permission profiles directly (#22795) ↗ openai/codex
  3. [3] app-server: stop returning thread permission profiles (#22792) ↗ openai/codex
  4. [4] telemetry: tag sandboxes from permission profiles (#22791) ↗ openai/codex
  5. [5] context: remove legacy permissions instructions helper (#22790) ↗ openai/codex
  6. [6] v6.38.0 ↗ openai/openai-node
  7. [7] release: 6.38.0 ↗ openai/openai-node
  8. [8] v2.37.0 ↗ openai/openai-python
  9. [9] release: 2.37.0 ↗ openai/openai-python

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

the week in review Codex hands model defaults to the server the month in review Agents API lands in both OpenAI SDKs
Full archive → Weekly reviews →

Quick answers

What shipped in OpenAI on May 16, 2026?
OpenAI shipped matching SDK updates across Node and Python while Codex finished moving its permissions model into production, cleaning up years of legacy sandbox code in the process. In total, 26 commits, 26 pull requests, and 4 releases landed.
Who contributed to OpenAI on May 16, 2026?
1 developer shipped this update, including stainless-app[bot].
What were the notable OpenAI updates?
Move memory prompt injection to app-server extension (#22841), core: construct test permission profiles directly (#22795), and app-server: stop returning thread permission profiles (#22792).
♥

$ log OpenAI --recent

Safety identifiers reach images, plugin approvals bind to contents 93 shipped Credential masking lands in Codex, Live gains Decisions across both SDKs 100 shipped Codex stabilizes environment tools, python adds residency options 135 shipped openai-python restores streaming connection reuse, browser and computer toolsets land in both SDKs 123 shipped

$ log --all # elsewhere on the wire

Alignment & Post-training verl rejects conflicting vLLM engine_kwargs, open-instruct adds MILES OPD behind a Megatron learner Interpretability nnsight 0.8.0 cuts import time from 7.6s to 1.6s

$ ls showcases/

Safety Evals AI Red Teaming Agent Security Interpretability Alignment & Post-training

$ 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.