站酷每日推荐日报

自动获取站酷（ZCOOL）首页热门设计作品（带链接、分类、亮点描述、趋势统计）

Security Scan

The skill's code and instructions match its stated purpose (scraping ZCOOL homepage items, generating descriptions, and saving a local report); it makes outbound HTTP requests and writes files locally but does not request unrelated credentials or perform suspicious actions.

Purpose & Capability

Name/description (daily ZCOOL recommendations) align with the provided script and example output. The included Python script performs HTML retrieval, classification, highlight generation, and saves a local report — all expected for this purpose.

Instruction Scope

SKILL.md simply instructs running the provided Python script. The script only fetches https://www.zcool.com.cn/ and individual work pages, parses titles/descriptions, classifies items, generates highlights, prints, and writes a text file under the output directory. It does not read other local files, environment variables, or external endpoints outside ZCOOL.

Install Mechanism

No install spec; this is instruction-plus-script. Dependencies are normal Python libraries (requests, beautifulsoup4) installable via pip, which is proportionate to the task. No downloads from unknown hosts or archive extraction are present.

Credentials

The skill requests no environment variables, secrets, or config paths. The absence of credentials is consistent with scraping public webpages.

Persistence & Privilege

Flags show no 'always: true' and default autonomous invocation behavior. The skill does not attempt to modify other skills or system-wide settings; it only writes its own output files to a local directory.

Assessment

This skill appears coherent and implements what it claims: it will make HTTP requests to zcool.com.cn and individual work pages, parse the HTML, and save a text file (zcool_daily/zcool_{date}.txt). Before installing or enabling autonomous runs, consider: (1) review the included scripts yourself — the code here is readable and not obfuscated; (2) be aware it will perform outbound network requests (ensure you are comfortable with that and that it complies with ZCOOL's site policies/robots.txt); (3) it writes files to the working directory (zcool_daily) — run in a directory or container where that is acceptable; (4) install dependencies in a virtualenv to avoid affecting system Python; (5) if you will allow the agent to invoke the skill autonomously, note that it will periodically perform network activity when invoked — schedule or restrict execution as needed. If you want extra assurance, run the script in an isolated environment (container/VM) the first time.

站酷每日热门作品推荐

功能说明

获取站酷（ZCOOL）首页的热门设计作品，输出格式：

• 标题带链接（可点击查看详情）
• 作品类型分类（AI绘画、插画、3D、动画、品牌设计等）
• 自动生成亮点描述
• 趋势统计（当日类型分布）

输出格式示例

详见 references/output-example.md

# 查看示例
cat references/output-example.md

亮点描述规则

• 针对热门作品标题有专门优化
• 根据关键词匹配自动生成
• 每个类型有默认描述作为兜底

使用方法

python3 scripts/zcool_daily.py --mode list

输出文件

作品列表保存至：zcool_daily/zcool_{date}.txt

依赖

• Python 3
• requests
• beautifulsoup4

安装：

pip install requests beautifulsoup4

支持的类型分类

AI绘画、插画、品牌设计、UI/UX、3D、动画、游戏、摄影、海报、手办、汽车/工业、电商、视频/MV