A native LLM client for iOS & Apple Watch. Run local GGUF models offline via llama.cpp, or connect to OpenAI/Claude/Gemini. Features local RAG, Model Context Protocol (MCP) tools, Siri Shortcuts, and cross-device sync. Built with Swift.
# Add to your Claude Code skills
git clone https://github.com/Eric-Terminal/ETOS-LLM-StudioGuides for using ai agents skills like ETOS-LLM-Studio.
Last scanned: 5/30/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-05-30T16:26:05.892Z",
"npmAuditRan": true,
"pipAuditRan": true
}See how ETOS-LLM-Studio compares with popular alternatives.
ETOS-LLM-Studio is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Eric-Terminal. A native LLM client for iOS & Apple Watch. Run local GGUF models offline via llama.cpp, or connect to OpenAI/Claude/Gemini. Features local RAG, Model Context Protocol (MCP) tools, Siri Shortcuts, and cross-device sync. Built with Swift. It has 203 GitHub stars.
Yes. ETOS-LLM-Studio passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/Eric-Terminal/ETOS-LLM-Studio" and add it to your Claude Code skills directory (see the Installation section above).
ETOS-LLM-Studio is primarily written in Swift. It is open-source under Eric-Terminal on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh ETOS-LLM-Studio against similar tools.
No comments yet. Be the first to share your thoughts!
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
一个运行在 iOS 和 Apple Watch 上的原生 AI / Agent 客户端。支持 OpenAI、Anthropic Claude、Google Gemini 与本机 GGUF / llama.cpp 模型,并提供会话级 Chat / Agent 模式、本地 Linux、Browser Agent、64 个 Apple 原生能力工具、MCP、Agent Skills、跨会话协作、本地 RAG 记忆、实时活动与双端同步。
English | 繁體中文 | 日本語 | Русский
在学校的日子挺无聊的,平时又总会冒出很多想问 AI 的问题。当时我嫌 App Store 上的 AI 应用要么贵得离谱,要么功能太残废(尤其是手表端),索性就自己动手搓了一个。
从最初那个只有 1,800 行代码、API Key 还要硬编码的简陋版本,到现在 971 个 Swift 源文件、362,394 行 Swift 代码(仅计算项目内 Swift,不把 llama.cpp、iSH 子模块和 VitePress 文档站依赖算进来)的工程,它确实已经长大了不少。虽然名字叫 "ETOS LLM Studio" 听着有点唬人,但它本质上还是我拿来探索大模型应用边界的试验场。
现在它已经不只是一个手表端 App 了:iOS 和 watchOS 都能管理云端模型、本地 GGUF 权重、工具、记忆、世界书和每日脉冲;Agent 还可以操作受控浏览器、调用 Apple 原生能力,或在 App 内置的本地 Linux 中运行命令与 Skill 脚本。两端数据通过内置同步引擎互通。
因为我平时主要还是用 Mac 和 Watch,iPhone 端偶尔会有一些还在继续打磨的边角,不过我会继续慢慢补齐。
.elsbackup、Cherry Studio、RikkaHub、Kelivo、ChatBox、ChatGPT conversations 等第三方会话,并可导出 PDF / Markdown / TXT。SFSpeechRecognizer 流式识别,iOS / watchOS 录音流程内嵌到聊天输入栏,支持实时转写、音频直发和识别结果回填输入框。getSystemTime 等能力,支持按来源和用途分组、聊天工具开关、审批策略、会话级启用、分类收纳与工具详情页。scripts/ 中的脚本;Skill 目录只读挂载,脚本版本与哈希随 Run 冻结,不能借运行中替换文件扩权。CustomJSTools 独立目录,创建前会执行脚本验证,可像普通工具一样启用、禁用和配置审批策略。ish-multiarch 嵌入最小 Alpine AArch64 RootFS,覆盖 iOS 与 watchOS。系统按实际操作懒安装,不会因为打开开关就启动,也不会静默安装 Python、Node.js 或编译工具;常用环境通过展示命令、软件源和存储影响的一键 recipe 由用户主动执行。linux_run、linux_shell、linux_process、交互式 PTY 终端、任务页、输出分页、取消和诊断。用户终端、Agent 命令与 Agent PTY 相互隔离,但共享授权后的 RootFS 与工作区;标准输出持续在后台收集,不依赖聊天页面保持打开。app://、linux:// 与 mount://,不会另造一套绕过审批的 Linux 文件 API。本地 stdio MCP 与 HTTP / SSE 服务器使用同一管理页、排序、工具开关和审批策略,并支持常见 mcpServers JSON 导入导出。character_book 等常见世界书格式。.elsbackup 上传与安全恢复流程。.elsbackup 快照导入、手表端全量导入、CloudKit 传输(含 APNs 静默推送触发后台同步)、iCloud Drive 备份导出/导入、启动备份、损坏自愈,以及通过 S3 兼容对象存储(AWS S3 / Cloudflare R2)签名上传快照、浏览远端快照并从云端下载恢复。@AppStorage,所有运行时配置走 GRDB 持久化、运行时缓存读取并以后台异步写入派发回主线程,避免主线程 I/O 与多设备配置漂移;支持旧版 UserDefaults 配置的一次性迁移。说实话,我最开始是想做免费软件的。 但 Apple Developer Program 每年 $99 的费用,对我一个学生来说确实有点吃力。
后来有位投资人帮我垫付了这笔钱,代价是我需要通过软件收费来偿还这笔投资(而且还要分成给他)。所以 App Store 版本象征性地收了一点费用,这就当是大家众筹帮我还债,顺便买个“不用每七天重签一次”的便利服务。
但是,开源是我的底线。
所以现在的规则很简单:
技术本该共享,我不希望因为几十块钱的门槛,挡住了同样对代码感兴趣的你。
@AppStorage,GRDB 持久化 + 运行时缓存 + 后台异步写入)local-llama-cpp 提供商libetos-llama.a, Accelerate / Metal(watchOS 运行期固定 CPU 路径)ish-multiarch、内置 Alpine AArch64 RootFS、Meson + Ninja 预编译 libiSHApple.a、PTY、动态挂载、guest 文件 API 与本地 stdio MCP