Kazuhiro Sera

New post on the OpenAI Developer Blog: how we use skills for open-source maintenance, from planning and coding to testing and release-readiness checks with GitHub Actions. Hope it's useful for your projects too 🙌 developers.openai.com/blog/skills-ag…

@Eduardopto @OpenAIDevs In this context, the end-to-end tests verify that there are no regressions in existing production API calls (yes, this could be somewhat specific to SDKs). Also, checking the full diff from the previous release tag helps us detect any breaking changes for existing SDK users.

@OpenAIDevs repeatable workflows break the moment your test environment drifts from prod. what's the verification step catching that before release, not after?

We use skills to maintain our Agents SDK repositories through repeatable workflows for verification, integration tests, release checks, and PR handoff. Here’s how it works: developers.openai.com/blog/skills-ag…

@coreyching @hirohasusan 嬉しいコメントをいただき、ありがとうございます！🙏

@hirohasusan ありがとう🙏🏼 @seratch can help provide more info

Ccdex + GPT-5.4でSRPGを作った例、画像も全てAIで生成 普段Codex使ってる人は「まあこのくらいなら作れるでしょうね」っていう感覚じゃないかな、今のCodexはそれくらい優秀

OpenAI Agents SDK TS 0.5.0: github.com/openai/openai-…

OpenAI Agents SDK Python 0.10.0: github.com/openai/openai-…

WebSocket mode for the OpenAI Responses API is now natively supported in the OpenAI Agents SDK! Check the release notes and the linked resources in the thread 🧵 developers.openai.com/api/docs/guide…

Skills cookbook: developers.openai.com/cookbook/examp…

Skills: developers.openai.com/api/docs/guide…

👋As you might expect, OpenAI Agents SDK now supports hosted shell tool + skills in an authentic way!

I became a developer because it allowed me to turn my ideas into reality... I joined Twilio because APIs allowed me to build more and have my code engage with the world... Grateful to be part of bringing Codex to the world to empower everyone to build. It's been amazing to see friends and family like my brother build and ship apps, something they weren't able to do before. We've reached a world where you can just build things! ❤️

If you're an OpenAI platform user, sorry for the wait! The long-awaited HITL feature is finally available in the Agents SDK for Python, just like in TS. If you skipped the SDK because this was missing, please give the latest version a try: github.com/openai/openai-…

Of course, the same one is available in TS SDK as well: github.com/openai/openai-…

We've added experimental support for using Codex as a tool within OpenAI Agents SDK apps: github.com/openai/openai-… If you're interested, please try it out and share your feedback with us!

Added some tiny skills to OpenAI Agents SDK repo for to boost productivity 🙂 github.com/openai/openai-…

2025 wasn’t about one launch - it was the year of stronger models, agent-native APIs and SoTA tooling! We shipped: - reasoning that converged (o1 → o3/o4-mini → GPT-5.2) - codex as a coding surface (GPT-5.2-Codex + CLI + web/IDE) - real multimodality (audio + realtime, images, video, PDFs) - agent-native building blocks (Responses API, Agents SDK, MCP) - open weight models (gpt-oss, gpt-oss-safeguard) And the curve moved fast (4o -> 5.2): GPQA 56.1% → 92.4% AIME 9.3% → 100% (!!) SWE-bench Verified 33.2 → 80.0 (!!!) Full recap: developers.openai.com/blog/openai-fo…

@gabrielchua @OpenAI @OpenAIDevs welcome!

Life update: I’m now at @OpenAI, working on Developer Experience across APAC. If you’re building in the region, I’d love to learn more. Tell me what sparks joy, what doesn't, and what you’d like to see next. @OpenAIDevs