Fetch website data
Skill lunrenyi/ai-agent.sh/.claude/skills/fetch-website-data
Fetch structured data from websites using x-cmd links dump. Use when the user needs to fetch or scrape website content, get README/repo info from GitHub/Gitee, or prepare LLM-friendly text from URLs.From its SKILL.md
npx -y skills add lunrenyi/ai-agent.sh --skill fetch-website-dataAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
- runs commandsInstructs the agent to run 3 commands, including `x-cmd links dump -l -c https://github.com/sigoden/aichat` and 2 more.
SKILL.md
1.0 KB, 273 tokens by cl100k_base, as published. Nobody here has run it
可以使用 x-cmd links dump 命令获取网站的 LLM 友好文本数据。
工作流程：
- 优先获取专为 LLM 优化的 llmstxt 格式
- 对于代码托管平台（GitHub、Gitee、Codeberg、GitLab），优先获取原始数据
- 回退方案：使用 elinks 文本浏览器提取内容
常用示例：
# 获取 GitHub 仓库信息（推荐，自动识别平台特性）
x-cmd links dump -l -c https://github.com/sigoden/aichat
# 仅获取原始数据（适用于代码仓库）
x-cmd links dump --llmstxt --codeurl <url>
# 抓取任意网页内容
x-cmd links dump <url>
参数说明：
-l, --llmstxt：输出专为 LLM 优化的文本格式-c, --codeurl：将代码链接转换为原始文件 URL（适合代码仓库）
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.