Wechat Article Formatter logo

Wechat Article Formatter

Community
iamzifei
wechat-article-formatter

Format markdown and publish to WeChat Official Account via bm.md rendering + WeChat official API 新手引导:输入 `/公众号排版 新手`、「这个怎么用」「第一次用」「能干嘛」时,走 SKILL.md 的〈新手上路〉,不要直接开始干活。

Overview

Publisheriamzifei
Repositorywechat-article-formatter-skill
Skill namewechat-article-formatter
Stars
77
Forks
11
Bundled files
2
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by iamzifei on GitHub. Read the source before you install it.

Installation

Install the Wechat Article Formatter AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/iamzifei/wechat-article-formatter-skill/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/iamzifei/wechat-article-formatter-skill.git \
  .claude/skills/wechat-article-formatter
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Wechat Article Formatter in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Wechat Article Formatter on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Wechat Article Formatter is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

WeChat Article Formatter & Publisher

Format markdown articles with bm.md styling and publish directly to WeChat Official Account drafts via the official API.

Overview

  1. Read markdown file (typically wechat.md from article directory)
  2. Upload local images to WeChat CDN via media/uploadimg API
  3. Replace local image paths with WeChat CDN URLs in markdown
  4. Render markdown to styled HTML via bm.md API with custom CSS
  5. Publish to WeChat draft via official draft/add API

CRITICAL: Do NOT use third-party publishing APIs (e.g. 微绿 wechat-publish). They strip inline CSS styles. Always use WeChat's official draft/add API directly.

Credentials

Read from ~/.env:

VariablePurpose
WECHAT_APP_IDWeChat Official Account AppID
WECHAT_APP_SECRETWeChat Official Account AppSecret

If not found, prompt the user to provide them and save to ~/.env.

Workflow

Step 1: Identify Input

The user provides a markdown file path (e.g. wechat.md in an article directory).

Detect the article directory to find:

  • promotion.md → extract digest/summary
  • _attachments/ → local images
  • Brand config from the conversation context → author name

Step 2: Get WeChat Access Token

python
params = urlencode({
    'grant_type': 'client_credential',
    'appid': WECHAT_APP_ID,
    'secret': WECHAT_APP_SECRET
})
# GET https://api.weixin.qq.com/cgi-bin/token?{params}

If error 40164 (IP not in whitelist), instruct user to add the IP at: mp.weixin.qq.com → 设置与开发 → 基本配置 → IP白名单.

Step 3: Upload Images to WeChat CDN

CRITICAL: Upload ALL images in the SAME session. WeChat CDN URLs from previous sessions may expire.

Use media/uploadimg API for inline content images:

python
# POST https://api.weixin.qq.com/cgi-bin/media/uploadimg?access_token={token}
# Content-Type: multipart/form-data
# Body: media=@image_file
# Response: {"url": "http://mmbiz.qpic.cn/..."}

Use material/add_material API for cover/thumb image:

python
# POST https://api.weixin.qq.com/cgi-bin/material/add_material?access_token={token}&type=image
# Content-Type: multipart/form-data
# Body: media=@cover_image
# Response: {"media_id": "...", "url": "..."}

For remote images (e.g. https://i.ibb.co/...), download first, then upload to WeChat CDN.

After uploading, replace all local/remote image paths in the markdown with WeChat CDN URLs before rendering.

Step 4: Render with bm.md

Read custom CSS from {{SKILL_DIR}}/styles/custom.css.

IMPORTANT: Write the JSON payload to a temp file and use curl -d @file to avoid shell escaping issues.

python
payload = json.dumps({
    "markdown": md_content,
    "markdownStyle": "green-simple",
    "codeTheme": "kimbie-light",
    "customCss": css_content,
    "enableFootnoteLinks": True,
    "openLinksInNewWindow": True,
    "platform": "wechat"
}, ensure_ascii=False)

# Write to /tmp/bm-payload.json, then:
# curl -s -X POST https://bm.md/api/markdown/render \
#   -H "Content-Type: application/json" \
#   -d @/tmp/bm-payload.json

The ensure_ascii=False is critical — without it, Chinese characters become \uXXXX escape sequences.

Step 5: Publish to WeChat Draft

python
draft_payload = json.dumps({
    "articles": [{
        "title": title,
        "author": author,                    # e.g. "在悉尼和稀泥"
        "thumb_media_id": thumb_media_id,     # from Step 3 material upload
        "content": html,                      # from Step 4 bm.md render
        "digest": digest,                     # from promotion.md, max ~60 Chinese chars
        "need_open_comment": 1,               # enable comments
        "only_fans_can_comment": 0,           # allow all users to comment
        "original_article_type": 1,           # declare as original content
        "reward_wording": "喜欢这篇文章,请我喝杯咖啡",  # enable tipping
        "creation_source_type": 1             # 个人观点,仅供参考
    }]
}, ensure_ascii=False).encode('utf-8')

# POST https://api.weixin.qq.com/cgi-bin/draft/add?access_token={token}

Default publish settings:

FieldValueDescription
need_open_comment1Enable comments
only_fans_can_comment0All users can comment
original_article_type1Declare as original
reward_wording"喜欢这篇文章,请我喝杯咖啡"Enable tipping
creation_source_type1个人观点,仅供参考

Digest extraction from promotion.md:

Look for ## 文章摘要 section, extract text, truncate to ~60 Chinese characters (WeChat limit is 120 bytes).

Author resolution (first match wins):

  1. User explicitly specified
  2. Brand config author field
  3. Brand config 品牌名称 field

Step 6: Report Result

WeChat Publishing Complete!

Title: {title}
Author: {author}
Digest: {digest}
Original: ✓ declared
Tipping: ✓ enabled
Comments: ✓ open to all
Images: {N} uploaded to WeChat CDN
media_id: {media_id}

→ Preview at: https://mp.weixin.qq.com (内容管理 → 草稿箱)

Markdown Content Order for WeChat

When building the WeChat markdown, elements should appear in this order:

1. 关注引导文字 (e.g. 👆 「关注」加「星标」...)
2. 封面图 (cover image)
3. --- 分隔线
4. 正文内容 (article body with inline images)
5. --- 分隔线
6. 作者介绍 (about author section)
7. --- 分隔线
8. 推广区块 (promo blockquote with image)
9. --- 分隔线
10. 互动引导 (engagement CTA)

The H1 title should be REMOVED from markdown — WeChat displays its own title from the title field.

Common Pitfalls

PitfallSolution
Chinese shows as \uXXXXUse ensure_ascii=False in ALL json.dumps() calls
Images not displayingUpload in same session; don't reuse old CDN URLs
Styles stripped (plain text)Use WeChat official draft/add API, NOT third-party APIs
40164 IP whitelist errorAdd current IP at mp.weixin.qq.com → 基本配置 → IP白名单
45004 digest too longKeep digest under 60 Chinese characters
40007 invalid media_idUpload cover via material/add_material first to get thumb_media_id
> [!NOTE] renders wrongConvert GitHub alerts to regular > blockquotes for WeChat

bm.md API Reference

Render: POST https://bm.md/api/markdown/render

ParameterTypeDefaultDescription
markdownstringrequiredMarkdown content
markdownStylestringayu-lightStyle ID (use green-simple)
codeThemestringkimbie-lightCode highlight theme
customCssstring""Custom CSS scoped to #bm-md
enableFootnoteLinksbooleantrueConvert links to footnotes
openLinksInNewWindowbooleantrueLinks open in new window
platformstringhtmlTarget: html, wechat, zhihu, juejin

Available styles: ayu-light, bauhaus, blueprint, botanical, green-simple, maximalism, neo-brutalism, newsprint, organic, playful-geometric, professional, retro, sketch, terminal

Styling Features

The custom CSS (styles/custom.css) provides:

  • Optima/Microsoft YaHei fonts
  • Green accent color (rgb(53, 179, 120))
  • H2 headings: white text on black background
  • Bold text: green color
  • Blockquotes: green left border
  • Responsive tables with alternating row colors
  • Code blocks with dark background

Dependencies

  • WeChat Official Account API credentials (~/.env)
  • bm.md rendering service (no auth required)
  • Custom CSS at {{SKILL_DIR}}/styles/custom.css

新手上路(用户不知道该输入什么时,走这里)

触发:/公众号排版 新手、「这个怎么用」「第一次用」「能干嘛」「带我走一遍」, 以及用户输入了技能名却没有给任何任务的时候。

这个模式的铁律:不假设、不索取。用户可能什么都没准备, 不要一上来就问他要文件、要 API key、要具体需求。按下面四步走:

一、先说清楚这是什么(三句话以内)

一句话:把 Markdown 排成公众号能直接用的样式。 公众号编辑器不认 Markdown,手动调格式很费时间, 这个技能把排版做完,还能换主题。

二、给编号选项,让他按回车就能继续

不要问开放式问题(「你想做什么?」对新手是负担)。给 3 个选项加一个默认:

想先看哪个?(直接回车 = 1)
  1. 先看看排出来长什么样(示例)
  2. 排我的这篇 Markdown
  3. 有哪些主题可以选

三、直接演示一遍,边做边解释

选完立刻做给他看,用示例数据,不需要他提供任何东西。 每做完一步,加一行「💡 刚才发生了什么」,一句话说明这步的意义。

四、毕业

演示完只问一个是非题:「要不要用你自己的文章真跑一遍?」 答是就进正常流程;答否就告诉他随时回来输 /公众号排版 新手。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Wechat Article Formatter AI skill do?

Format markdown and publish to WeChat Official Account via bm.md rendering + WeChat official API 新手引导:输入 `/公众号排版 新手`、「这个怎么用」「第一次用」「能干嘛」时,走 SKILL.md 的〈新手上路〉,不要直接开始干活。

Why use Wechat Article Formatter on TypingMind?

Because you install it once and use it with any model. Wechat Article Formatter is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Wechat Article Formatter in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/iamzifei/wechat-article-formatter-skill/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Wechat Article Formatter?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Wechat Article Formatter?

As many as you like. As long as a model supports skills, you can use Wechat Article Formatter with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Wechat Article Formatter AI skill free?

It is published on GitHub by iamzifei. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇