[Idea] A community-maintained i18n foundation & Chinese (zh-CN) localization for the global "vibe coding" community

[zhenweilai]

Summary & Motivation

First of all, huge respect and thank you to Theo, Julius, and the entire Ping Labs team for building T3 Code. The experience of orchestrating agents across git worktrees, unified harnesses (Claude Code, Antigravity, Codex, etc.), and seamless remote access has completely transformed how development feels in the AI era.

I’m reaching out on behalf of the fast-growing Chinese developer and builder community. Over the past few months, "vibe coding" (AI-first, agent-driven development) has exploded in popularity across China. Solo creators, product managers, independent hackers, and professional engineers alike are rapidly embracing agent-native workflows.

Among these users, T3 Code has quickly become one of the most celebrated and sought-after tools. However, the single most common barrier preventing tens of thousands of enthusiastic users from adopting T3 Code as their daily driver is the lack of multi-language support—specifically Simplified Chinese (zh-CN).

We know from earlier discussions (such as #1463) that the core team has limited capacity and cannot take on the maintenance burden of translations. We completely understand and respect this: shipping core agent features, ensuring runtime stability, and maintaining architectural velocity must come first.

The goal of this proposal is not to ask the core team to write or maintain translations, but to explore introducing a minimal, extensible i18n foundation so that the community can carry 100% of the ongoing localization effort.

---

Why This Matters: The Chinese Vibe Coding Wave

1. A Massive, Highly Engaged User Base:
   China currently represents one of the largest and most active communities of AI-first builders and "vibe coders" in the world. Countless developers and creators are eager to use T3 Code to drive their projects.

2. Lowering Cognitive Friction for Everyday Use:
   While experienced programmers can read English documentation, "vibe coding" empowers a much wider spectrum of creators—designers, PMs, and junior builders. For non-native English speakers, having native menus, settings, permissions dialogs, and git conflict summaries dramatically reduces hesitation and cognitive fatigue, making agents much safer and more natural to control.

3. Preventing Community Fragmentation:
   Because demand is so strong, community members have already started creating unofficial forks (e.g. ZhiweiXiao98/t3code) and packaging third-party releases just to get Chinese UI. While this showcases real demand, maintaining disconnected forks fragments the community and prevents users from seamlessly receiving official nightlies, upstream fixes, and T3 Connect updates. An upstream-supported i18n foundation keeps everyone on the official mainline.

---

A Zero-Overhead Architecture Proposal for the Core Team

We want this to be zero-burden for the core team:

1. Compile-Time, Typed String Catalogs:

   • en-US remains the authoritative source of truth.
   • Use a lightweight typed dictionary pattern (e.g. t('settings.appearance.theme') with compile-time safety and graceful fallback to English).
   • Core team members continue writing English strings as normal—no extra translation steps required when shipping new features.

2. 100% Community-Maintained Translations:

   • The Chinese community is ready to maintain, test, and update the zh-CN dictionary via lightweight PRs or a platform like Crowdin/Weblate.
   • Core maintainers don't need to review translation nuances; they only need to merge verified community catalog updates.

3. Incremental Scope:

   • Focus initially on high-value UI surfaces: Settings panel, Permission & Confirmation dialogs, Menu navigation, Status badges.
   • Technical values, CLI commands, model identifiers, and provider names remain in raw English to prevent confusion.

---

Open Question to Maintainers

Would the team be open to guidance or a minimal RFC PR that introduces just the baseline i18n architecture/hooks, with the explicit agreement that translation authoring and maintenance will be handled entirely by the community?

Thank you again for creating such a visionary tool. We would love to collaborate to make T3 Code accessible to millions of builders worldwide!

[juliusmarminge]

Note

Grok responding on behalf of Julius.

Thanks for the thoughtful write-up. This is the same request as #6789 (community-led Simplified Chinese localization on a minimal i18n foundation), so I'm closing this one as a duplicate to keep the conversation in one place. Please add your points there.

Related work you may want to follow: #15179 (opt-in English / Simplified Chinese catalogs on V2), and the open PRs #16087 (shared i18n foundation) and #16673 (opt-in Simplified Chinese settings and navigation).