← Back to Skills Marketplace
skill everyday
by
do_while_true
· GitHub ↗
· v1.0.0
· MIT-0
126
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install skill-everyday
Description
每天抓取 Clawhub 热门技能，深入分析并生成报告。每次执行获取一个未分析过的热门技能，避免重复。
Usage Guidance
该技能总体一致且用途明确，但在安装/运行前请注意：
- 此脚本会在本地创建 data/ 并写入报告（包括固定的 LATEST.md），若不希望将生成内容保存在某路径请调整或在隔离目录运行。
- 脚本会尝试访问父级的 skills/ 目录并列出目标技能的文件名（不会读取文件内容），如果宿主机上有敏感工程或私有代码，建议在隔离环境（容器或干净目录）中运行。
- 运行前需安装 Node.js 与 Playwright，并执行 npx playwright install chromium（会下载浏览器二进制，占用下载/磁盘资源）。
- 源自 unknown（未提供 homepage/作者信任信息），若你信任该来源则可以使用；否则建议先人工审阅 runner.mjs 并在受控环境中试运行，或直接把脚本拷贝到受控目录并禁用本地目录检查（注释掉读取父目录相关代码）。
总结：功能与所需操作相符、未见隐蔽外发或凭据窃取行为；主要注意点是本地文件列表访问和 Playwright/Chromium 的安装成本。
Capability Analysis
Type: OpenClaw Skill
Name: skill-everyday
Version: 1.0.0
The skill-everyday bundle is a tool designed to scrape the Clawhub website for popular OpenClaw skills and generate local analysis reports. It uses Playwright in scripts/runner.mjs to intercept API calls from clawhub.ai (specifically via wry-manatee-359.convex.cloud) and fetch skill metadata. The script also checks for the presence of local skill directories in the parent folder to include file listings in its reports, a behavior that is transparently documented in SKILL.md. There is no evidence of data exfiltration, unauthorized remote execution, or malicious prompt injection.
Capability Assessment
Purpose & Capability
技能名/描述与实现基本一致：定期抓取热门技能并生成本地报告。唯一超出直觉的点是脚本会尝试访问宿主环境中上一级的 skills/ 目录（targetSkillDir），列出本地技能文件名以便把本地代码作为额外上下文纳入报告——这与“深入分析”可以解释得通，但会读取并展示本地文件列表（不读文件内容）。
Instruction Scope
SKILL.md 与 runner.mjs 的指示一致：用 Playwright 打开 clawhub.ai、拦截并读取 Convex API 响应以获取榜单，打开技能详情页抓取标题/描述，并在本地写入 data/reports 和 data/analyzed.json。没有看到把本地或抓取到的数据发往未经申明的远端服务；唯一硬编码的外部端点是 Convex 的请求 URL（wry-manatee-359.convex.cloud），用于从网页拦截榜单数据。
Install Mechanism
无安装规范（instruction-only），但仓内包含 package.json（声明 playwright 依赖）和运行脚本；README 明确建议 npm install + npx playwright install chromium。风险/影响可见：Playwright 会下载 Chromium 等大文件，且需在运行环境中安装 Node.js 与浏览器二进制。
Credentials
不要求任何环境变量或外部凭据（primaryEnv: none），这是合理且比例适当。但脚本会访问文件系统：创建 data/、写入 reports 与 analyzed.json，并尝试列出父目录下目标技能目录的文件名（可能暴露本地项目结构）。没有看到要求或使用不相关的密钥/令牌。
Persistence & Privilege
没有设置 always:true；技能仅在自身目录下持久化状态（data/、data/reports/LATEST.md、data/analyzed.json）。没有修改其他技能配置或系统范围设置，agent-autonomy 保持默认（可被 agent 调用）。
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install skill-everyday - After installation, invoke the skill by name or use
/skill-everyday - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of skill-everyday: daily analysis and reporting on trending Clawhub skills.
- Automatically avoids duplicates by tracking analyzed skills in data/analyzed.json.
- No platform login or API Key required; data is scraped from public Clawhub pages using Playwright and Chromium.
- Generates and saves structured reports for each new skill in data/reports/.
- Simple command-line execution: node scripts/runner.mjs.
Metadata
Frequently Asked Questions
What is skill everyday?
每天抓取 Clawhub 热门技能，深入分析并生成报告。每次执行获取一个未分析过的热门技能，避免重复。 It is an AI Agent Skill for Claude Code / OpenClaw, with 126 downloads so far.
How do I install skill everyday?
Run "/install skill-everyday" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is skill everyday free?
Yes, skill everyday is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does skill everyday support?
skill everyday is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created skill everyday?
It is built and maintained by do_while_true (@autoxj); the current version is v1.0.0.
More Skills