← 返回 Skills 市场
S2 Atmos Perception
作者
MilesXiang
· GitHub ↗
· v1.0.9
· MIT-0
164
总下载
0
收藏
0
当前安装
8
版本数
在 OpenClaw 中安装
/install s2-atmos-perception
功能描述
S2-SP-OS Atmos Radar. Real-time meteorological and space weather (NOAA) perception organ for AI Agents.
安全使用建议
This skill appears coherent and limited to fetching public weather/AQI and NOAA space‑weather data. Before installing, note that: (1) it will send any user-entered location string to third‑party APIs (Open‑Meteo and NOAA), so treat sensitive location data accordingly; (2) provenance is unknown (no homepage), so if you require strict supply-chain assurance run the script locally in an isolated environment or review the file yourself (atmos.py is included and small); and (3) network access is required and the tool will make outbound HTTPS requests — if that is undesirable, do not enable the skill. Otherwise the skill's permissions and behavior are proportionate to its stated purpose.
功能分析
Type: OpenClaw Skill
Name: s2-atmos-perception
Version: 1.0.9
The skill bundle is a legitimate meteorological and space weather data fetcher. The Python script (atmos.py) uses standard libraries to query public APIs from Open-Meteo and NOAA, and the instructions in SKILL.md include explicit safety constraints preventing the agent from automatically executing physical device commands.
能力评估
Purpose & Capability
Name and description (real-time meteorology + NOAA space weather) align with the included Python script and SKILL.md. Required binary is only python3 and no credentials or config paths are requested — consistent with calling public weather/AQI and NOAA endpoints.
Instruction Scope
SKILL.md instructs the agent to request the user's location and to run the bundled script; the script sends the provided location string to public APIs (Open-Meteo geocoding, Open-Meteo weather/air-quality, and NOAA SWPC). This is expected for the stated purpose but does transmit the user-provided location to third-party services.
Install Mechanism
No install spec — instruction-only with a bundled Python file. No downloads from arbitrary URLs or archive extraction. Low installation risk.
Credentials
The skill requires no environment variables, credentials, or config paths. It only needs network access to public APIs, which is proportionate to fetching weather and space-weather data.
Persistence & Privilege
always is false and the skill does not request persistent system modifications or elevated privileges. allowed-tools includes exec (to run python3), which is reasonable for running the included script.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install s2-atmos-perception - 安装完成后，直接呼叫该 Skill 的名称或使用
/s2-atmos-perception触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.9
s2-atmos-perception 1.0.9
- Replaced s2_sensor_array.py with atmos.py, streamlining codebase.
- SKILL.md revised: clarified privacy rule, simplified agent/admin instructions, and updated terminology (e.g., "environmental_insights" instead of "s2_actuation_suggestions").
- Updated tags and metadata; removed redundant privacy-focused language and unnecessary tags.
- Improved agent guidance: no automatic device actions, stronger user-consent requirements.
- General documentation and feature summary improvements.
v1.0.8
S2-Atmos-Perception 1.0.8
- Refined privacy policy: explicit city request required for all queries.
- Simplified user instructions for AI Agents and Human Admins.
- Command-line usage updated: now use atmos.py instead of s2-atmos-perception.py.
- Clarified execution flow for actuation suggestions and prohibited arbitrary script calls.
- Updated description and documentation for accuracy and conciseness.
v1.0.7
Minor updates and cleanup for clarity and consistency.
- Added "Space Weather" to tags for improved discoverability.
- Updated the metadata section: removed unused emoji for simplification.
- Standardized all script file names to use hyphens (e.g., s2-atmos-perception.py).
- Improved wording for instructions to be more concise and uniform.
- Reduced use of emojis and excess formatting for a cleaner document.
v1.0.6
- Major documentation overhaul with improved clarity and structure in SKILL.md.
- Files README.md, manifest.json, and requirements.txt removed.
- Enhanced description of privacy-first principles and agent instructions.
- Expanded details about input/output tensors and execution workflow.
- Clearer separation of instructions for AI agents and human admins.
v1.0.3
- No changes detected in this version.
- Skill remains unchanged; behavior and instructions are identical to the previous version.
v1.0.2
- Privacy-first update: Auto geolocation via IP is removed; now requires explicit user-supplied location.
- Improved instructions: You must ask the user for their city if unknown—no guessing or inferring from IP.
- Revised execution flow: Only run the radar after confirming the location from user input or conversation history.
- Minor clarifications to tensor interpretation and actuation steps for transparency and user consent.
v1.0.1
- Updated handling of S2 actuation suggestions to clarify execution: do not run arbitrary bash commands.
- Now requires extracting protocol, s2_element, device_id, and intent from the recommended_tensor object for actuation.
- Provided an explicit (step-by-step) procedure for proposing and executing physical world changes.
- Enhanced emphasis on accuracy and eliminating hallucinated commands when acting on suggestions.
v1.0.0
S2-Atmos-Perception 1.0.0
- Introduces S2 Atmos Radar for atmospheric, air quality, and space weather sensing.
- Provides command-line interface syntax for automatic or city-specific environmental perception.
- Details interpretation rules for weather, air quality (AQI), indoor CO₂ levels, and geomagnetic storms.
- Explains handling of S2 Actuation Suggestions, including how to communicate and execute recommended actions.
- Establishes structured responses based on JSON tensor outputs for improved feedback and action.
元数据
常见问题
S2 Atmos Perception 是什么？
S2-SP-OS Atmos Radar. Real-time meteorological and space weather (NOAA) perception organ for AI Agents. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 164 次。
如何安装 S2 Atmos Perception？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install s2-atmos-perception」即可一键安装，无需额外配置。
S2 Atmos Perception 是免费的吗？
是的，S2 Atmos Perception 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
S2 Atmos Perception 支持哪些平台？
S2 Atmos Perception 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 S2 Atmos Perception？
由 MilesXiang（@spacesq）开发并维护，当前版本 v1.0.9。
推荐 Skills