agentleFS
kakafengkakafeng/wechat-publisher-2.0/SKILL.md

wechat-publisher-2.0

Skill1 starsChanged 6 months agoMITView on GitHub

微信公众号文章发布器。Markdown 写稿 → 一键发到公众号草稿箱。 内置选题诊断、调研提纲、去AI味检查、封面图生成、质量检查、飞书双写。 触发词:发公众号、发布文章、公众号草稿箱、markdown转公众号。

  • Reads credentials
  • Installs packages

The file

Files
SKILL.md

wechat-publisher-2.0 — 公众号发布器

一句话

Markdown 写文章 → 一键发到公众号草稿箱,零外部项目依赖。

为什么选这个 Skill

  • 全流水线:选题诊断 → 调研提纲 → 去AI味 → 质量检查 → 草稿箱,一步到位
  • 零外部依赖:不依赖 wechat-allauto-gzh 或任何已失的外部项目
  • 纯 PIL 封面:无需 AI 绘图 API,零成本生成封面
  • SQLite 持久化:文章台账、选题库、发布日志一应俱全
  • 飞书双写:发布后自动同步到多维表格(可选)

依赖安装

cd skills/wechat-publisher-2.0
pip install -r requirements.txt

环境变量

变量必填
WECHAT_APP_ID✅
WECHAT_APP_SECRET✅
TAVILY_API_KEY可选(调研搜索)
STEPFUN_API_KEY可选(调研备用)
FEISHU_APP_ID可选(飞书双写)
FEISHU_APP_SECRET可选(飞书双写)

使用方式

方式一:完整流水线(推荐)

# 一键走完全流程
python3 pipeline.py publish \
    --title "人工智能对旅游行业的影响" \
    --md article.md \
    --category travel

# 仅诊断不发布
python3 pipeline.py publish --title "xxx" --md article.md --diagnose-only

方式二:分步使用

# 诊断选题
python3 pipeline.py diagnose "AI Agent 对旅游业的影响"

# 调研+提纲
python3 pipeline.py research "AI Agent 对旅游业的影响" --outline

# 检查/修复 AI 味
python3 pipeline.py humanizer-check article.md
python3 pipeline.py humanizer-fix article.md --output cleaned.md

# 发布
python3 wechat_publisher_2.py publish --title "文章标题" --md article.md --auto

方式三:编程调用

from wechat_publisher_2 import Publisher

pub = Publisher()
result = pub.publish(
    title   = "2026 年,组团社为什么越来越慌",
    markdown= pathlib.Path("article.md").read_text(),
    digest  = "AI 正在从三个层面消解组团社的生存根基…",
    author  = "冯浪",
    theme   = "default",
    category= "travel",
    auto_submit=False,
)

核心能力

Markdown → WeChat HTML

支持标题、引用、粗体/斜体/删除线、代码块、列表、分隔线、链接,以及 57 个 HTML Anything 排版模板。

选题诊断(5 维度)

维度权重
读者规模25%
实操价值25%
个人经历20%
传播潜力20%
竞争密度10%

去 AI 味检查

自动扫描并替换 20+ 中英文 AI 常用表达(crucial, leverage, "在当今...时代" 等)。

封面图自动生成

基于 PIL 随机生成,无需 AI 绘图 API。色板按 travel / tech / culture / business / marine 分类。

SQLite 台账

表内容
articles文章正文、HTML、状态、MediaID
topics选题库、状态、优先级
publish_logs每次发布操作日志

文件结构

wechat-publisher-2.0/
├── SKILL.md                 # 本文件
├── README.md                # 项目说明(徽章、安装、demo)
├── LICENSE                  # MIT License
├── requirements.txt         # Python 依赖
├── .gitignore               # Git 忽略规则
├── wechat_publisher_2.py    # 核心发布逻辑(~1200行)
├── pipeline.py              # 增强流水线(诊断+调研+去AI味)
├── topic_diagnose.py        # 选题诊断(5维度打分)
├── topic_research.py        # 调研辅助 + 去AI味检查
└── themes/                  # Markdown → HTML 主题(运行时生成)

详细文档、故障排查和版本历史请参阅 README.md。

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.