pr
免费Create or update a pull request
中文简介正在准备中。当前显示原文说明。
Use when asked about Turbo (Turbo Drive, Turbo Frames, Turbo Streams, ActionCable-broadcasted streams) or Stimulus (controllers, targets, actions, values, outlets) — including "how do I do X with Turbo/Stimulus", debugging a Turbo Stream broadcast, or wiring up a Stimulus controller. Always consult the official Hotwire docs rather than answering from memory alone, since Turbo/Stimulus APIs and idioms have shifted across versions and this project has no SPA framework — all interactivity goes through Hotwire.
中文简介正在准备中。当前显示原文说明。
实测评分 未实测
查看安装方法安装前可以查看交给 AI 代理的指令内容。
This project (HostedGPT) uses Hotwire (Turbo + Stimulus) for all frontend interactivity — no SPA
framework. Streaming chat updates arrive via Turbo Stream broadcasts over ActionCable, rendered
from server-side partials (see app/views/messages/_message.html.erb and CLAUDE.md). Trained
knowledge about Turbo/Stimulus can be stale or blend behavior across versions — check the official
docs before relying on specifics.
Use WebFetch against the sources above when:
append, prepend, replace, update,
remove, before, after, refresh) or how targeting (target=/targets=) works.data-turbo-* attributes, or navigation
interception behaves.event->controller#method).CLAUDE.md.Skip the lookup for questions answered faster by reading this codebase directly — e.g. how this
app's own streaming job builds and broadcasts updates is answered by
app/services/ai_backend/, GetNextAIMessageJob, and the _message.html.erb partials, not the
Hotwire docs.
The Hotwire docs describe the general framework; this project's own streaming architecture (see
CLAUDE.md's "Message generation flow") governs how it's actually used here — e.g. partial
updates are broadcast roughly every 100ms during streaming and only persisted once the stream
finishes. Confirm Turbo/Stimulus behavior against the docs, but follow this repo's existing
patterns for where controllers and broadcasts live.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Create or update a pull request
中文简介正在准备中。当前显示原文说明。
Catch up on the team's merges since you last checked in
中文简介正在准备中。当前显示原文说明。
Address unresolved PR review comments from Copilot and teammates
中文简介正在准备中。当前显示原文说明。
Review the branch against project guidelines, fix what it finds, strip non-load-bearing comments, and keep .claude/rules fresh
中文简介正在准备中。当前显示原文说明。
Use when asked about any Rails-specific topic — ActiveRecord, routing, controllers, views, mailers, jobs, Action Cable, Action Text, Active Storage, migrations, validations, callbacks, associations, caching, security, or Rails internals — including "how do I do X in Rails", debugging Rails behavior, or checking whether an API/option still exists or changed in a given Rails version. Always consult the official guides and API docs rather than answering from memory alone, since trained knowledge can be stale or wrong about version-specific details.
中文简介正在准备中。当前显示原文说明。
Build and maintain Ruby or Rails applications with the RubyLLM AI framework. Use for chats, agents, tools, structured output, media generation, transcription, OCR, moderation, embeddings, reranking, Rails integration, and RubyLLM upgrades; not for contributing to the framework itself.
中文简介正在准备中。当前显示原文说明。