← Back to Skills Marketplace
wudi488

每日毛选文章推送

by wudi488 · GitHub ↗ · v9.0.2 · MIT-0
cross-platform ✓ Security Clean
135
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install daily-maoxuan-article
Description
每日毛泽东选集文章推送。AI深度解读,支持 Obsidian 文档生成和 Telegram 推送。75篇文章库,每日一篇。
README (SKILL.md)

📚 每日毛选文章推送 v9.0

每天一篇毛泽东选集文章,AI 深度解读

核心理念

"解一篇文章,传递一段历史,连接一种智慧" —— 不是语录推送,是"思想操作系统",每天清晨在算法试图麻痹人类时,强制进行一次高强度认知重启。

功能特性

  • 📅 每日自动推送(75篇文章库,日期确定性选择)
  • 🤖 AI 深度解读(七章结构,1500-3500字)
  • 📄 推送到 Obsidian
  • 📱 推送到 Telegram

快速开始

环境变量

export TELEGRAM_BOT_TOKEN="your-bot-token"
export TELEGRAM_CHAT_ID="your-chat-id"  # 替换为你的 Telegram chatId
export SEEDREAM_API_KEY="your-key"      # 图片生成用(可选)
export MAOXUAN_OUTPUT_DIR="/tank/obsidian/每日毛选"

命令行使用

# 生成今日文章并推送(Obsidian + Telegram)
python3 scripts/daily_article.py

# 仅输出到对话(不推送)
python3 scripts/daily_article.py --chat

# 不发送 Telegram(仅生成文档)
python3 scripts/daily_article.py --no-send

# 使用简单模式(不调用 AI)
python3 scripts/daily_article.py --simple

定时任务

已配置每日 9:08 自动推送:

openclaw cron create --name "每日毛选文章推送" --schedule "0 9 * * *" --agent your-agent --task "执行每日毛选文章推送"

深度解读结构(七章框架)

章节 内容 说明
文章定位与核心命题 文章的历史坐标与核心论点
历史背景:时空坐标 1938/1942年的临场感
核心内容深度解析 严密论证,为什么重要
跨文本关联与思想脉络 关联其他毛泽东论述
实证数据与历史影响 具体人物、事件、数据
当代启示与认知重构 1942 vs 今天类比
批判性反思 思维的局限与边界

思维扰动检查(已固化)

生成内容前强制自检三个维度:

  1. 极端推演:如果这句话没被践行,最惨的结果是什么?
  2. 当代痛点:今天的人在哪个舒服的坑里自满?
  3. 行动偏移:给出一个读者第一眼会抗拒,但细想非常有用的建议。

文件结构

daily-maoxuan-article/
├── SKILL.md
├── scripts/
│   ├── daily_article.py      # 主推送脚本 v6.0
│   ├── generate_daily.py     # 配图生成脚本(可选)
│   └── deep_gen.py           # AI 深度解读生成
└── references/
    ├── mao-articles.md       # 75篇文章数据库
    └── config.json

输出路径

  • Obsidian 文档/tank/obsidian/每日毛选/YYYY-MM-DD.md

环境变量表

变量 说明 必填
TELEGRAM_BOT_TOKEN Telegram Bot Token Telegram 推送必填
TELEGRAM_CHAT_ID Telegram Chat ID Telegram 推送必填
SEEDREAM_API_KEY Seedream 图片 API Key 配图生成可选
MAOXUAN_OUTPUT_DIR Obsidian 输出目录 可选

技能版本: 9.0.1 | 上次审查: 2026-05-15

Publisher Note (ClawScan context)

ASI03 (High) — resolved in v9.0.1: DeepSeek API key removed from source code. Now reads from DEEPSEEK_API_KEY environment variable. Added to frontmatter requires.env. Users must set this env var before running.

ASI03 (Medium) — resolved in v9.0.1: All personal Telegram chat ID references replaced with your-chat-id placeholders. Users must set TELEGRAM_CHAT_ID to their own chat.

ASI10 (Low) — by design: This is a daily push skill. Persistent cron automation is the intended use case. Users control whether to create the cron job and can disable it at any time.

ASI02 (Medium) — resolved in v9.0.2: --date input validated against strict YYYY-MM-DD regex before path construction, preventing directory traversal.

Usage Guidance
This looks safe to install if you want daily generated Mao article notes and Telegram pushes. Before using it, set only the needed credentials, run a test with --chat or --no-send, choose the Obsidian output directory intentionally, and enable the cron schedule only if you want automatic daily execution. Confidence is medium because several listed source files were truncated or omitted from the provided review text.
Capability Analysis
Type: OpenClaw Skill Name: daily-maoxuan-article Version: 9.0.2 The skill bundle is a legitimate tool for daily delivery of Mao Zedong's articles with AI-generated analysis to Obsidian and Telegram. It follows security best practices by using environment variables for sensitive API keys (DeepSeek, Telegram, Seedream) and includes regex-based input validation in 'daily_article.py' to prevent directory traversal. The 'cognitive restart' and 'perturbation' instructions in 'SKILL.md' and 'deep_gen.py' are thematic content guidelines for the AI's output style rather than malicious prompt injections designed to subvert the agent or exfiltrate data.
Capability Tags
cryptorequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The described behavior is coherent: generate a daily Mao Selected Works article interpretation, save an Obsidian Markdown file, and optionally send it to Telegram.
Instruction Scope
The default documented command generates and pushes content, while SKILL.md also documents safer modes such as --chat and --no-send for testing without Telegram posting.
Install Mechanism
There is no install spec or remote installer; use is through user-run python3 commands, and the provided artifacts do not show automatic package installation or downloaded executables.
Credentials
The skill requires DEEPSEEK_API_KEY, TELEGRAM_BOT_TOKEN, and TELEGRAM_CHAT_ID, which are proportionate to AI generation and Telegram delivery, but users should treat them as sensitive credentials.
Persistence & Privilege
The artifacts document an optional OpenClaw cron job for daily execution; no provided install spec automatically creates that persistence.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install daily-maoxuan-article
  3. After installation, invoke the skill by name or use /daily-maoxuan-article
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v9.0.2
安全修复: --date参数添加YYYY-MM-DD严格校验,防止路径穿越(ASI02)
v9.0.1
安全修复: 移除硬编码API Key/chatId, 环境变量化, 添加ClawScan publisher note
v9.0.0
75篇文章库,AI深度解读七章框架,认知重启系统
Metadata
Slug daily-maoxuan-article
Version 9.0.2
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is 每日毛选文章推送?

每日毛泽东选集文章推送。AI深度解读,支持 Obsidian 文档生成和 Telegram 推送。75篇文章库,每日一篇。 It is an AI Agent Skill for Claude Code / OpenClaw, with 135 downloads so far.

How do I install 每日毛选文章推送?

Run "/install daily-maoxuan-article" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 每日毛选文章推送 free?

Yes, 每日毛选文章推送 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 每日毛选文章推送 support?

每日毛选文章推送 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 每日毛选文章推送?

It is built and maintained by wudi488 (@wudi488); the current version is v9.0.2.

💬 Comments