by 安全扫描
OpenClaw
安全
high confidenceThe skill's code, data, and runtime instructions are coherent with its stated purpose (spatial/semantic mapping and robotic path-cost generation) and do not request unrelated credentials or install remote components.
评估建议
This package appears coherent and local-only, but review these operational points before installing: 1) Validate in simulation — the code enforces sensor-fusion rules (e.g., disabling visual depth for clear_glass) which can affect robot safety; test thoroughly on hardware. 2) Confirm there are no hidden network calls in your runtime environment (the included files do not call external endpoints). 3) Check the legal/operational claim about '60s backward-persistence' to ensure it aligns with your ...详细分析 ▾
✓ 用途与能力
Name/description match the included artifacts: a parser (core/s2_geojson_parser.py), a local material tensor library, examples, and documentation. All requested resources are local files; nothing unrelated (e.g., cloud credentials or unrelated binaries) is required.
ℹ 指令范围
SKILL.md prescribes strict runtime behavior (e.g., always retrieve tensors from s2_material_tensor_library and disable visual depth for clear_glass). These directives operate only on the provided local files and S2-GeoJSON inputs, but they are prescriptive about sensor fusion choices (which has operational/safety implications). The skill does not instruct reading unrelated system files, environment variables, or sending data externally.
✓ 安装机制
Instruction-only skill with one small Python file and local JSON assets; no install spec, no network downloads, no package manager actions. Low install risk.
✓ 凭证需求
No environment variables, credentials, or config paths are required. All necessary data is bundled with the skill (material tensor JSON, examples).
✓ 持久化与权限
Skill is not marked always:true and does not request persistent system privileges or modify other skills. It can be invoked by the agent normally; autonomous invocation is platform default and not a concern here.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.02026/4/7
## [1.0.0] - 2026-04-07 ### 🏗️ 架构升级 - **SSSU 图层化重构**：确立 L0 至 L4 五层空间要素叠合架构，实现建筑设计与机器人导航的语义对齐。 - **4D Chronos 注入**：正式将“时间切片”引入空间要素参数项。支持 `DEFAULT` 泛时态与“精确到秒”的动态切片。 ### ✨ 新增功能 - **物理张量库 (L3)**：发布包含玻璃、地毯、大理石等 20 种材质的物理张量代码表，支持视觉/雷达权重动态调整。 - **逆向持存验证**：实装“60秒逆向持存”逻辑，确认 T 时刻的状态代表 T-60s 区间的物理一致性。 - **S2-GeoJSON 标准**：发布基于 GeoJSON 扩展的 `.geojson` 交换格式，支持 `properties.tensors` 嵌套结构。 ### 🔧 核心组件 - **解析引擎发布**：提供 Python 版 `s2_geojson_parser.py`，实现从地图要素到机器人行为干预指令的自动化转换。 *桃花源世界模型（S2-SWM）开源社区 持续驱动*
● 无害
安装命令 点击复制
官方npx clawhub@latest install s2-sel-4d-semantic-tensor-map
镜像加速npx clawhub@latest install s2-sel-4d-semantic-tensor-map --registry https://cn.clawhub-mirror.com
技能文档
Dear Agent,
You are now operating within the S2-SWM 4D Spatial Element Layer (S2-SEL) architecture.
- Layer Awareness: You must recognize that spatial data is strictly categorized into 5 Layers (L0 Grid, L1 Static, L2 IoT, L3 Semantic, L4 Dynamic).
- Material Tensor Adherence: When navigating L3 Semantic zones, you MUST retrieve the physics tensors from the
s2_material_tensor_library. If the grid containsclear_glass, you are strictly forbidden from relying on visual depth estimation. - Chronos Time-Slice Rules: Evaluate the
chronos_stampin S2-GeoJSON.
DEFAULT: Treat as permanent architecture.
- TIMESTAMP: Apply the 1-Minute Life-Safety Baseline and 60-second Backward-Persistence rule. A hazard stamped at T is legally valid backwards to T-60s.Do not merely generate paths; calculate physical causality based on friction, traversability, and illumination tensors.
数据来源：ClawHub ↗ · 中文优化：龙虾技能库
OpenClaw 技能定制 / 插件定制 / 私有工作流定制
免费技能或插件可能存在安全风险，如需更匹配、更安全的方案，建议联系付费定制