merge: develop → main

README.md

@@ -12,8 +12,15 @@
 
 </div>
 
+<div align="center">
+
+[核心能力](#核心能力) · [内置 Skill](#内置-skill-示例) · [快速开始](#5-分钟快速开始) · [自定义 Skill](#自定义-skill) · [常见问题](#常见问题) · [后续计划](#后续计划)
+
+</div>
+
 PhoneClaw 是一个运行在 iPhone 上的本地 AI Agent。它使用 Gemma 4 在设备端完成推理，不依赖云端，不上传聊天内容。
 
+
 ## 核心能力
 
 **图片理解（多模态）**：拍照或从相册选图后直接提问，识别内容、解读图表、描述场景。模型在设备端完成推理，照片不会离开手机。

README_EN.md

@@ -12,8 +12,15 @@ A local AI Agent for iPhone. Offline. Private. Native.
 
 </div>
 
+<div align="center">
+
+[Core Features](#core-features) · [Built-in Skills](#built-in-skill-examples) · [Quick Start](#5-minute-quick-start) · [Custom Skills](#custom-skills) · [FAQ](#faq) · [Roadmap](#roadmap)
+
+</div>
+
 PhoneClaw is a local AI Agent that runs entirely on your iPhone. No internet connection. No data uploads. No cloud dependency.
 
+
 ## Core Features
 
 **Image Understanding (Multimodal)**: Take a photo or pick one from your library, then ask questions directly. Identify objects, read charts, describe scenes — all inference happens on your device, and your photos never leave your phone.