Search

PythonReactSEOPDF
OpenClaw

Newspaper Brief

Newspaper Brief

eisonme0 source installs☆ 0 Repository stars

Installation

Follow the repository instructions for this skill and your agent. Review the files and dependencies before adding it to your workflow.

SkillHub CLI

skillhub install eisonme/newspaper-brief

Inspect the source and compatibility requirements, then follow its installation steps.

Give your agent the source link and ask it to inspect instructions and dependencies before explaining the installation steps.

View source ↗Guides

Skill instructions

Newspaper Brief

把整理结果从“聊天流”切成“版面流”,先产出适合手机阅读的报纸风长图。

默认工作方式

优先使用这套流程:

  1. 读用户给的原始材料(文章、聊天记录、链接摘要、会议纪要)
  2. 先整理成结构化要点
  3. 再渲染成报纸风长图

如果用户已经给了现成摘要/要点,也可以直接跳到第 3 步。

触发词

以下中文说法都应直接进入本技能:

  • 报纸模式
  • 做成报纸图
  • 按报纸样式输出

也兼容这些自然说法:

  • 整理成简报图
  • 做成报纸样式
  • 用报纸风输出
  • 给我一张报纸风长图

v1 范围

  • 仅做 手机版长图
  • 先输出 HTML,再截图为 PNG
  • 默认风格:黑白报纸 + 轻米色纸张背景 + 少量强调线
  • 适合 Telegram 直接查看,不追求打印排版

数据准备

优先整理成下面这种结构,再交给脚本:

{
  "paper_name": "紫音简报",
  "issue": "2026-03-07",
  "title": "AI 输出界面值得重做",
  "subtitle": "把聊天流改造成版面流,比单纯堆更长回答更有效",
  "summary": "一句导语,放在标题下方。",
  "highlights": [
    "先给 3-5 条 TL;DR",
    "正文使用 2-4 个版块分栏",
    "手机端版式必须优先保证可读性"
  ],
  "quote": "AI 时代,真正该重做的可能不是模型,而是输出界面。",
  "sections": [
    {
      "heading": "核心判断",
      "body": "这里放正文段落。支持多段文本。"
    },
    {
      "heading": "产品建议",
      "bullets": ["建议一", "建议二", "建议三"]
    }
  ],
  "footer_note": "来源:群聊讨论|整理:紫音"
}
原始材料模式

当用户只给原文/聊天记录时,先由代理自己完成提炼:

  • 先写标题
  • 再写一句导语
  • 提炼 3-5 条 highlights
  • 拆成 2-4 个 sections
  • 最后补一句 footer
标题规则(v1.1)
  • 标题最多两行
  • 不允许用 ... / …… 省略截断
  • 默认优先 重写成更短、更像报纸标题的表达
  • 只有在仍然接近边界时,才轻微缩小字号

不要把未经整理的大段原文直接硬塞进模板,除非用户明确要求“原汁原味”。

脚本

1) 生成 HTML
python skills/newspaper-brief/scripts/render_newspaper.py \
  --input skills/newspaper-brief/examples/ai-interface-sample.json \
  --html output/newspaper-brief/demo.html \
  --normalized output/newspaper-brief/demo.normalized.json
2) 可选:顺手导出...

Source information

Source
github
Repository
openclaw/skills
Path
skills/eisonme/newspaper-brief
Version
—
Commit
9199b3cf569322cc230ab34ee464ad7309c1cca5
Classification source
rules
Status
active
Visibility
public

These signals come from indexed source analysis and may lag behind changes. Check the current source files.

Extracted content

SKILL.md
# Newspaper Brief

把整理结果从“聊天流”切成“版面流”,先产出适合手机阅读的报纸风长图。

## 默认工作方式

优先使用这套流程:

1. 读用户给的原始材料(文章、聊天记录、链接摘要、会议纪要)
2. 先整理成结构化要点
3. 再渲染成报纸风长图

如果用户已经给了现成摘要/要点,也可以直接跳到第 3 步。

## 触发词

以下中文说法都应直接进入本技能:

- `报纸模式`
- `做成报纸图`
- `按报纸样式输出`

也兼容这些自然说法:

- `整理成简报图`
- `做成报纸样式`
- `用报纸风输出`
- `给我一张报纸风长图`

## v1 范围

- 仅做 *手机版长图*
- 先输出 **HTML**,再截图为 **PNG**
- 默认风格:黑白报纸 + 轻米色纸张背景 + 少量强调线
- 适合 Telegram 直接查看,不追求打印排版

## 数据准备

优先整理成下面这种结构,再交给脚本:

```json
{
  "paper_name": "紫音简报",
  "issue": "2026-03-07",
  "title": "AI 输出界面值得重做",
  "subtitle": "把聊天流改造成版面流,比单纯堆更长回答更有效",
  "summary": "一句导语,放在标题下方。",
  "highlights": [
    "先给 3-5 条 TL;DR",
    "正文使用 2-4 个版块分栏",
    "手机端版式必须优先保证可读性"
  ],
  "quote": "AI 时代,真正该重做的可能不是模型,而是输出界面。",
  "sections": [
    {
      "heading": "核心判断",
      "body": "这里放正文段落。支持多段文本。"
    },
    {
      "heading": "产品建议",
      "bullets": ["建议一", "建议二", "建议三"]
    }
  ],
  "footer_note": "来源:群聊讨论|整理:紫音"
}
```

### 原始材料模式

当用户只给原文/聊天记录时,先由代理自己完成提炼:

- 先写标题
- 再写一句导语
- 提炼 3-5 条 highlights
- 拆成 2-4 个 sections
- 最后补一句 footer

### 标题规则(v1.1)

- 标题最多两行
- 不允许用 `...` / `……` 省略截断
- 默认优先 **重写成更短、更像报纸标题的表达**
- 只有在仍然接近边界时,才轻微缩小字号

不要把未经整理的大段原文直接硬塞进模板,除非用户明确要求“原汁原味”。

## 脚本

### 1) 生成 HTML

```bash
python skills/newspaper-brief/scripts/render_newspaper.py \
  --input skills/newspaper-brief/examples/ai-interface-sample.json \
  --html output/newspaper-brief/demo.html \
  --normalized output/newspaper-brief/demo.normalized.json
```

### 2) 可选:顺手导出...

Related skills

OpenClaw

x402hub

capncoconut

Register, communicate, and earn on the x402hub AI agent marketplace. Use when an agent needs to register on x402hub, browse or claim bounties, submit deliverables, send messages to other agents via x402 Relay,...

OpenClaw

claw

capevace

Real-time event bus for AI agents. Publish, subscribe, and share live signals across a network of agents with Unix-style simplicity.

OpenClaw

captchas-openclaw

captchasco

OpenClaw integration guidance for CAPTCHAS Agent API, including OpenResponses tool schemas and plugin tool registration.