← Back to Skills Marketplace
137
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install yuntu
Description
获取并返回最新的实时卫星云图。当用户提到"云图"、"卫星云图"、"实时云图"、"天气图"等关键词时触发。 此技能会自动下载最新云图，并发给用户。
Usage Guidance
This skill appears to do exactly what it says: download a FY4B satellite image, OCR the timestamp, compress and cache the image, and print the local path for the agent to send. Before installing, consider: (1) you will need to install system packages (Tesseract) and Python libs—follow trusted package sources for those; (2) the Windows instructions suggest downloading a Tesseract language file from GitHub raw—verify the URL before downloading; (3) the script writes images to ~/.openclaw/satellite_cache and will make them available to the agent (the agent may upload them when it uses the <qqmedia> tag); (4) if you prefer tighter control, run the script manually in a sandbox to verify behavior first. Overall the skill is coherent and there are no obvious signs of malicious behavior.
Capability Analysis
Type: OpenClaw Skill
Name: yuntu
Version: 1.0.0
The skill is designed to fetch and display real-time satellite cloud maps from the National Satellite Meteorological Center (nsmc.org.cn). The code in get_satellite.py performs legitimate image downloading, OCR processing to extract timestamps, and local file management within a dedicated cache directory (~/.openclaw/satellite_cache), with no evidence of data exfiltration, unauthorized execution, or malicious intent.
Capability Assessment
Purpose & Capability
Name/description match the included code and SKILL.md. The script downloads a single published FY4B image, runs OCR to extract the timestamp, compresses and caches the image, and prints the local path for the agent to include in a reply. No unrelated services, credentials, or binaries are requested.
Instruction Scope
SKILL.md instructs installing tesseract and Python libs and running the provided script; the script's behavior (download, OCR, compress, save path) matches those instructions. The instructions do not ask the agent to read unrelated files or exfiltrate other data. The agent is expected to embed the local image path (via <qqmedia>)—this is consistent with the skill's goal.
Install Mechanism
There is no built-in install spec; the README-style SKILL.md suggests installing OS packages and pip dependencies (requests, pillow, pytesseract). These are common and proportionate. The only external file download referenced is an optional Tesseract chi_sim traineddata from GitHub raw, which is an expected source for language data but should be verified by the user.
Credentials
The skill requires no environment variables, credentials, or config paths. It only performs network access to a single public image URL and writes cached JPG files under the user's home (~/.openclaw/satellite_cache), which is proportionate to functionality.
Persistence & Privilege
The skill does not request always: true or other elevated persistence. It writes its own cache files under ~/.openclaw and does not modify other skill or system configurations.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install yuntu - After installation, invoke the skill by name or use
/yuntu - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
yuntu
- 卫星云图下载：支持关键词“云图”“卫星云图”“实时云图”等自动触发。
- 自动下载并发送风云4B最新卫星云图。
- 回复内容包含格式化拍摄时间和云图图片。
- 大小限制，不能上传打包好的二进制文件，安装OCR库容易出错，拍摄时间错误没办法
Metadata
Frequently Asked Questions
What is 卫星云图?
获取并返回最新的实时卫星云图。当用户提到"云图"、"卫星云图"、"实时云图"、"天气图"等关键词时触发。 此技能会自动下载最新云图，并发给用户。 It is an AI Agent Skill for Claude Code / OpenClaw, with 137 downloads so far.
How do I install 卫星云图?
Run "/install yuntu" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 卫星云图 free?
Yes, 卫星云图 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 卫星云图 support?
卫星云图 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 卫星云图?
It is built and maintained by wuliao8 (@wuliao8); the current version is v1.0.0.
More Skills