基于 SOC 职业分类
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/supremeb/Oniva-ai --skill extract-structured-data命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | extract_structured_data |
| description | Extract structured data (tables, lists) from HTML content using CSS selectors. |
| type | tool |
| category | web |
| tool | {"module":"app.skills.extract_structured_data.tool","function":"extract_structured_data","async":true} |
| metadata | {"author":"oniva","version":"1.0.0"} |
| audit-level | basic |
Parse HTML and extract tables and lists.
result = await extract_structured_data(html="<html>...</html>", selector="table.data")
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
html | string | Yes | - | HTML content to parse |
selector | string | No | None | CSS selector to focus on specific elements |
{
"tables": [
[["Header1", "Header2"], ["Row1Col1", "Row1Col2"]]
],
"lists": [
["Item 1", "Item 2", "Item 3"]
],
"tables_count": 1,
"lists_count": 1
}
result = await extract_structured_data(html=page_html)
for table in result["tables"]:
print(table)
result = await extract_structured_data(
html=page_html,
selector="div.financial-data"
)