update README & PRD

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2025 nanobot contributors
+Copyright (c) 2025-2026 OpenCane contributors
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal
@@ -18,4 +18,4 @@ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.
+SOFTWARE.

README.md

@@ -9,6 +9,14 @@ Language: English | [切换到简体中文](#简体中文)
 OpenCane is an AI backend runtime for smart-cane scenarios, focused on an end-to-end loop:
 device connectivity -> realtime conversation -> visual lifelog -> digital tasks -> safety and observability.
 
+### Product Snapshot
+
+- Product goal: build an OpenClaw experience for visually impaired users
+- Interaction carrier: smart-cane hardware
+- Human-device interaction: voice + buttons (input), voice + vibration (feedback)
+- Core loop: hardware access -> realtime conversation -> visual memory -> digital tasks -> safety/observability
+- Runtime form: backend service with control APIs and multi-modem adapter layer
+
 ### Architecture
 
 ![OpenCane Architecture](docs/assets/opencane-arch.png)
@@ -22,6 +30,16 @@ device connectivity -> realtime conversation -> visual lifelog -> digital tasks
 - Digital task execution: create, query, cancel, offline pushback, and retry
 - Control APIs: device registration/binding, operation dispatch, and runtime status queries
 
+### Functional Scenarios
+
+- On-the-go voice assistance: press-to-talk interaction for quick answers while walking.
+- Forward risk check: photo + voice queries to identify nearby obstacles and hazards.
+- Short-term memory recall: ask about recent moments via timeline and semantic retrieval.
+- Digital task delegation: turn voice intents into trackable tasks with status updates.
+- Device health and offline alerts: detect heartbeat/connection anomalies and notify users.
+- Emergency help trigger (SOS): minimal-action emergency trigger with immediate confirmation.
+- Digital Journal (emotion-first): multimodal models understand daily photo sequences and voice context to form long-term memory; it can be replayed with personalized, podcast-like narration and shared as text with family and friends.
+
 ### Technical Highlights
 
 - Layered architecture: `adapter / runtime / agent / api / storage / safety`
@@ -95,6 +113,14 @@ Thanks to HKUDS for the original engineering base.
 
 OpenCane 是一个面向智能盲杖场景的 AI 后端运行时，聚焦“设备接入 -> 实时对话 -> 图像记忆 -> 数字任务 -> 安全与观测”的完整闭环。
 
+### 产品功能摘要
+
+- 产品目标：打造“给视障人士使用的 OpenClaw”
+- 交互载体：智能盲杖硬件
+- 人机交互：语音 + 按钮输入，语音 + 震动反馈
+- 核心闭环：设备接入 -> 实时对话 -> 图像记忆 -> 数字任务 -> 安全与观测
+- 系统形态：后端运行时 + 控制 API + 多蜂窝模组适配层
+
 ### 架构图
 
 ![OpenCane 架构图](docs/assets/opencane-arch.png)
@@ -108,6 +134,16 @@ OpenCane 是一个面向智能盲杖场景的 AI 后端运行时，聚焦“设
 - 数字任务执行：任务创建、状态查询、取消、离线回推与重试
 - 控制面 API：设备注册绑定、设备指令下发、运行状态查询
 
+### 功能场景
+
+- 行走中即时问答：按键触发语音提问，低延迟语音回复，不打断通行节奏。
+- 前方风险确认：拍照 + 语音询问，识别附近障碍与风险并给出提示。
+- 短时记忆回溯：支持“刚才发生了什么”的时间线与语义检索回答。
+- 数字任务代办：将语音需求转成可追踪任务，并持续回报状态。
+- 设备异常与离线提醒：心跳或连接异常时及时告警，恢复后可继续会话。
+- 紧急求助触发（SOS）：通过最少动作触发求助，并立即得到确认反馈。
+- 数字日记（情感优先）：多模态大模型理解每天的照片序列与语音形成长期记忆；可用个性化语音像播客一样回放，也可生成文本分享给亲人朋友。
+
 ### 技术特性
 
 - 分层架构：`adapter / runtime / agent / api / storage / safety`

case/README.md

@@ -0,0 +1,13 @@
+# Case Assets
+
+This directory stores demo GIF assets used for quick visual previews.
+
+- `code.gif`: code/tool workflow preview
+- `memory.gif`: memory/lifelog behavior preview
+- `schedule.gif`: task scheduling preview
+- `search.gif`: search/retrieval preview
+
+Notes:
+
+- Keep filenames lowercase and descriptive.
+- Prefer `.gif` only for short looped demonstrations.