Install
openclaw skills install s2-acoustic-perceptionS2-SP-OS Acoustic Radar. Edge-delegated zero-shot classification with Ephemeral Privacy and strict LAN-only network enforcement. / S2-SP-OS 语义声学雷达。本地边缘零样本分类,采用“阅后即焚”与严格的局域网物理隔离保障绝对隐私。
openclaw skills install s2-acoustic-perceptionWelcome to the S2 Acoustic Perception Client. Built on the principle of Ephemeral Privacy (阅后即焚) and Strict LAN Isolation (局域网绝对隔离). This radar listens to the environment and sends a short audio slice only to a verified private IP address (RFC 1918). It returns semantic meaning (e.g., "music", "dog_barking") and instantly drops the raw audio from RAM.
基于**“阅后即焚”与“局域网物理隔离”**的隐私理念。本雷达仅允许将短时音频切片发送至受验证的内网私有 IP(RFC 1918标准)。大脑返回语义标签后,原生音频在客户端内存中被立刻销毁。
--edge-ip to ensure data NEVER leaves your local network router. Public IPs are strictly blocked. / 代码级校验 IP 地址,公网 IP 将被直接熔断,确保数据绝对不出网关。Phase 1: Authorized Acoustic Read / 阶段 1:空间授权读取 Ensure the user grants consent. The Edge IP MUST be a private local address (e.g., 192.168.x.x, 10.x.x.x). / 确保用户授权,Edge IP 必须为私有内网地址。
python3 acoustic.py --mode read --edge-ip 192.168.1.100 --zone living_room --grid x2_y3 --consent-granted true
Phase 2: Action on Insights / 阶段 2:解析与行动
If audio_event is Human conversation, respect the user's privacy. Do not ask what they are saying.
If the event is pet_distress or glass_shattering, execute offline_linkage_suggestions to trigger security or vision adapters.