WeWrite — WeChat AI Publishing Skill

WeWrite — 微信公众号AI创作Skill

Skill by ara.so — Daily 2026 Skills collection.

WeWrite is a full-pipeline AI skill for producing WeChat Official Account (公众号) articles end-to-end: hotspot fetching → topic selection → writing → SEO → AI image generation → formatting → draft box publishing. It runs as a Claude Code skill (via

SKILL.md

) but every component also works standalone.

该Skill由ara.so开发——属于Daily 2026 Skills系列。

WeWrite是一款面向微信公众号文章的全流程AI Skill，提供端到端的创作支持：热点抓取→选题→写作→SEO优化→AI图片生成→格式排版→草稿箱发布。它作为Claude Code Skill运行（基于

SKILL.md

），同时每个组件也可独立使用。

Installation

安装步骤

As a Claude Code Skill

作为Claude Code Skill安装

bash

undefined

bash

undefined

Clone the repo

git clone https://github.com/oaker-io/wewrite.git ~/.claude/skills/wewrite

Or copy into an existing project

cp -r wewrite ~/.claude/skills/wewrite

undefined

cp -r wewrite ~/.claude/skills/wewrite

undefined

Python Dependencies

Python依赖安装

bash

cd wewrite
pip install -r requirements.txt

bash

cd wewrite
pip install -r requirements.txt

Configuration

配置设置

bash

cp config.example.yaml config.yaml

Edit

config.yaml

:

yaml

wechat:
  appid: "${WECHAT_APPID}"         # WeChat Official Account App ID
  secret: "${WECHAT_SECRET}"       # WeChat Official Account Secret

image_gen:
  provider: "doubao"               # "doubao" or "openai"
  doubao_api_key: "${DOUBAO_API_KEY}"
  openai_api_key: "${OPENAI_API_KEY}"

output_dir: "./output"

Set environment variables instead of hardcoding secrets:

bash

export WECHAT_APPID="wx1234567890abcdef"
export WECHAT_SECRET="your_secret_here"
export DOUBAO_API_KEY="your_doubao_key"
export OPENAI_API_KEY="sk-..."

bash

cp config.example.yaml config.yaml

编辑

config.yaml

：

yaml

wechat:
  appid: "${WECHAT_APPID}"         # 微信公众号App ID
  secret: "${WECHAT_SECRET}"       # 微信公众号密钥

image_gen:
  provider: "doubao"               # 可选"doubao"或"openai"
  doubao_api_key: "${DOUBAO_API_KEY}"
  openai_api_key: "${OPENAI_API_KEY}"

output_dir: "./output"

通过环境变量配置敏感信息，避免硬编码：

bash

export WECHAT_APPID="wx1234567890abcdef"
export WECHAT_SECRET="your_secret_here"
export DOUBAO_API_KEY="your_doubao_key"
export OPENAI_API_KEY="sk-..."

Triggering the Full Pipeline

触发全流程创作

Once installed as a Claude Code skill, simply say:

用 demo 的配置写一篇公众号文章

Claude will execute all steps automatically using

clients/demo/style.yaml

as the client profile.

You can also specify a client:

用 clients/tech-blog 的风格，围绕今日热点写一篇公众号文章，发布到草稿箱

安装为Claude Code Skill后，只需输入：

用 demo 的配置写一篇公众号文章

Claude会自动执行所有步骤，使用

clients/demo/style.yaml

作为客户配置文件。

你也可以指定特定客户配置：

用 clients/tech-blog 的风格，围绕今日热点写一篇公众号文章，发布到草稿箱

Pipeline Steps & Scripts

流程步骤与脚本

1. Fetch Hotspots

1. 热点抓取

Scrapes real-time trending topics from Weibo, Toutiao, and Baidu.

bash

python3 scripts/fetch_hotspots.py --limit 20
python3 scripts/fetch_hotspots.py --limit 10 --json   # JSON output

Output example:

json

[
  {"rank": 1, "title": "DeepSeek R2 发布", "source": "weibo", "heat": 98200},
  {"rank": 2, "title": "A股科技板块大涨", "source": "baidu", "heat": 75300}
]

从微博、头条和百度抓取实时热门话题。

bash

python3 scripts/fetch_hotspots.py --limit 20
python3 scripts/fetch_hotspots.py --limit 10 --json   # 输出JSON格式

输出示例：

json

[
  {"rank": 1, "title": "DeepSeek R2 发布", "source": "weibo", "heat": 98200},
  {"rank": 2, "title": "A股科技板块大涨", "source": "baidu", "heat": 75300}
]

2. SEO Keyword Analysis

2. SEO关键词分析

Queries Baidu and 360 search suggestions to score keywords.

bash

python3 scripts/seo_keywords.py "AI大模型" "科技股"
python3 scripts/seo_keywords.py --json "ChatGPT" "人工智能"

Python usage:

python

from scripts.seo_keywords import analyze_keywords

results = analyze_keywords(["AI大模型", "大语言模型", "GPT-5"])
for kw in results:
    print(f"{kw['keyword']}: score={kw['score']}, volume={kw['estimated_volume']}")

查询百度和360搜索建议，对关键词进行评分。

bash

python3 scripts/seo_keywords.py "AI大模型" "科技股"
python3 scripts/seo_keywords.py --json "ChatGPT" "人工智能"

Python调用示例：

python

from scripts.seo_keywords import analyze_keywords

results = analyze_keywords(["AI大模型", "大语言模型", "GPT-5"])
for kw in results:
    print(f"{kw['keyword']}: score={kw['score']}, volume={kw['estimated_volume']}")

3. Topic Selection

3. 选题生成

Claude reads

references/topic-selection.md

and generates 10 candidate topics scored on:

热度 (trending heat)
契合度 (client fit)
差异化 (differentiation)

Claude读取

references/topic-selection.md

，生成10个候选选题，并基于以下维度评分：

热度 (trending heat)
契合度 (client fit)
差异化 (differentiation)

4. Framework & Writing

4. 框架与写作

Claude reads

references/frameworks.md

(5 frameworks) and

references/writing-guide.md

(de-AI style rules), then writes the article adapted to the client's tone.

Claude读取

references/frameworks.md

（5种写作框架）和

references/writing-guide.md

（去AI化写作规则），然后根据客户语气调整文章内容。

5. AI Image Generation

5. AI图片生成

bash

undefined

bash

undefined

Cover image (recommended 900×383)

封面图（推荐尺寸900×383）

python3 toolkit/image_gen.py
--prompt "科技感封面，蓝色光线，未来感"
--output output/cover.png
--size cover

Inline content image

内文配图

python3 toolkit/image_gen.py
--prompt "程序员在办公室工作，现代风格插画"
--output output/img1.png
--provider openai


Python usage:

```python
from toolkit.image_gen import generate_image

path = generate_image(
    prompt="AI机器人与人类握手，科技感插画",
    output_path="output/cover.png",
    size="cover",          # "cover" (900x383) or "content" (800x600)
    provider="doubao"      # "doubao" or "openai"
)
print(f"Generated: {path}")

python3 toolkit/image_gen.py
--prompt "程序员在办公室工作，现代风格插画"
--output output/img1.png
--provider openai


Python调用示例：

```python
from toolkit.image_gen import generate_image

path = generate_image(
    prompt="AI机器人与人类握手，科技感插画",
    output_path="output/cover.png",
    size="cover",          # 可选"cover"(900x383)或"content"(800x600)
    provider="doubao"      # 可选"doubao"或"openai"
)
print(f"Generated: {path}")

6. Formatting — Markdown → WeChat HTML

6. 格式转换——Markdown转微信HTML

WeChat requires inline styles. The converter handles this automatically.

bash

undefined

微信公众号要求内联样式，转换器会自动处理这一点。

bash

undefined

Preview in browser

在浏览器中预览

python3 toolkit/cli.py preview article.md --theme professional-clean

Available themes

查看可用主题

python3 toolkit/cli.py themes


Python usage:

```python
from toolkit.converter import MarkdownConverter
from toolkit.theme import load_theme

theme = load_theme("tech-modern")
converter = MarkdownConverter(theme=theme)

with open("article.md") as f:
    markdown_content = f.read()

html = converter.convert(markdown_content)

python3 toolkit/cli.py themes


Python调用示例：

```python
from toolkit.converter import MarkdownConverter
from toolkit.theme import load_theme

theme = load_theme("tech-modern")
converter = MarkdownConverter(theme=theme)

with open("article.md") as f:
    markdown_content = f.read()

html = converter.convert(markdown_content)

html is WeChat-ready with all inline styles

生成的html是符合微信要求的内联样式格式

undefined

undefined

7. Publish to WeChat Draft Box

7. 发布到微信草稿箱

bash

python3 toolkit/cli.py publish article.md \
  --cover output/cover.png \
  --title "2026年AI大模型最新进展" \
  --author "科技观察"

Python usage:

python

from toolkit.publisher import WeChatPublisher
from toolkit.wechat_api import WeChatAPI

api = WeChatAPI(appid=os.environ["WECHAT_APPID"], secret=os.environ["WECHAT_SECRET"])
publisher = WeChatPublisher(api=api)

bash

python3 toolkit/cli.py publish article.md \
  --cover output/cover.png \
  --title "2026年AI大模型最新进展" \
  --author "科技观察"

Python调用示例：

python

from toolkit.publisher import WeChatPublisher
from toolkit.wechat_api import WeChatAPI

api = WeChatAPI(appid=os.environ["WECHAT_APPID"], secret=os.environ["WECHAT_SECRET"])
publisher = WeChatPublisher(api=api)

Upload cover image first

先上传封面图

media_id = api.upload_image("output/cover.png")

Push to draft box

推送至草稿箱

draft_id = publisher.create_draft( title="2026年AI大模型最新进展", content=html_content, # inline-styled HTML from converter cover_media_id=media_id, author="科技观察", digest="本文盘点2026年大模型最新进展..." # summary/excerpt ) print(f"Draft created: {draft_id}")

undefined

draft_id = publisher.create_draft( title="2026年AI大模型最新进展", content=html_content, # 来自转换器的内联样式HTML cover_media_id=media_id, author="科技观察", digest="本文盘点2026年大模型最新进展..." # 文章摘要 ) print(f"Draft created: {draft_id}")

undefined

8. Fetch Article Stats (回填数据)

8. 获取文章数据（回填）

bash

python3 scripts/fetch_stats.py --article-id "your_article_id"

bash

python3 scripts/fetch_stats.py --article-id "your_article_id"

9. Learn from Manual Edits

9. 从人工编辑中学习

bash

python3 scripts/learn_edits.py \
  --original output/draft.md \
  --edited output/final.md \
  --client demo

Extracts style rules from diffs and appends them to the client's playbook.

bash

python3 scripts/learn_edits.py \
  --original output/draft.md \
  --edited output/final.md \
  --client demo

从差异中提取风格规则，并追加到客户的规则手册中。

Client Configuration

客户配置

Each client lives in

clients/{name}/style.yaml

:

yaml

undefined

每个客户的配置存放在

clients/{name}/style.yaml

：

yaml

undefined

clients/my-tech-blog/style.yaml

name: "我的科技博客" industry: "科技/AI" topics:

"人工智能"
"大模型应用"
"编程技术" tone: "专业严谨，偶尔幽默，面向中级开发者" theme: "tech-modern" avoid:
"过度营销语言"
"绝对化表述" wechat: appid: "${WECHAT_APPID}" secret: "${WECHAT_SECRET}"


Create a new client:

```bash
mkdir clients/my-client
cp clients/demo/style.yaml clients/my-client/style.yaml

name: "我的科技博客" industry: "科技/AI" topics:

"人工智能"
"大模型应用"
"编程技术" tone: "专业严谨，偶尔幽默，面向中级开发者" theme: "tech-modern" avoid:
"过度营销语言"
"绝对化表述" wechat: appid: "${WECHAT_APPID}" secret: "${WECHAT_SECRET}"


创建新客户配置：

```bash
mkdir clients/my-client
cp clients/demo/style.yaml clients/my-client/style.yaml

Edit style.yaml for your client

编辑style.yaml适配你的客户需求

---

---

Themes

主题样式

Theme	Style
`professional-clean`	Clean professional (default)
`tech-modern`	Tech-forward blue/purple gradient
`warm-editorial`	Warm editorial tones
`minimal`	Minimal black/white

bash

python3 toolkit/cli.py themes          # list all themes with previews
python3 toolkit/cli.py preview article.md --theme warm-editorial

Custom theme (YAML):

yaml

undefined

主题	风格
`professional-clean`	简洁专业风格（默认）
`tech-modern`	科技感蓝紫渐变
`warm-editorial`	温暖编辑风格
`minimal`	极简黑白风格

bash

python3 toolkit/cli.py themes          # 列出所有主题及预览
python3 toolkit/cli.py preview article.md --theme warm-editorial

自定义主题（YAML格式）：

yaml

undefined

toolkit/themes/my-theme.yaml

name: my-theme body: font-family: "'PingFang SC', sans-serif" font-size: "16px" color: "#333" line-height: "1.8" h2: color: "#1a73e8" font-weight: "bold" border-left: "4px solid #1a73e8" padding-left: "10px" blockquote: background: "#f0f4ff" border-left: "3px solid #4285f4" padding: "12px 16px" color: "#555"

---

name: my-theme body: font-family: "'PingFang SC', sans-serif" font-size: "16px" color: "#333" line-height: "1.8" h2: color: "#1a73e8" font-weight: "bold" border-left: "4px solid #1a73e8" padding-left: "10px" blockquote: background: "#f0f4ff" border-left: "3px solid #4285f4" padding: "12px 16px" color: "#555"

---

Full Pipeline — Python Orchestration

全流程Python编排

python

import subprocess
import json
import os
from toolkit.converter import MarkdownConverter
from toolkit.theme import load_theme
from toolkit.publisher import WeChatPublisher
from toolkit.wechat_api import WeChatAPI
from toolkit.image_gen import generate_image

python

import subprocess
import json
import os
from toolkit.converter import MarkdownConverter
from toolkit.theme import load_theme
from toolkit.publisher import WeChatPublisher
from toolkit.wechat_api import WeChatAPI
from toolkit.image_gen import generate_image

1. Fetch hotspots

1. 抓取热点

result = subprocess.run( ["python3", "scripts/fetch_hotspots.py", "--limit", "20", "--json"], capture_output=True, text=True ) hotspots = json.loads(result.stdout)

2. SEO analysis on top topics

2. 对热门话题进行SEO分析

from scripts.seo_keywords import analyze_keywords top_titles = [h["title"] for h in hotspots[:5]] seo_scores = analyze_keywords(top_titles)

3. (Claude selects topic, writes article — handled by SKILL.md)

3. (Claude负责选题和写作——由SKILL.md处理)

After Claude produces article.md:

当Claude生成article.md后：

4. Generate cover

4. 生成封面图

cover_path = generate_image( prompt="科技感封面图，蓝色渐变，数字化未来", output_path="output/cover.png", size="cover", provider=os.environ.get("IMAGE_PROVIDER", "doubao") )

5. Convert Markdown → WeChat HTML

5. 转换Markdown为微信HTML

theme = load_theme("tech-modern") converter = MarkdownConverter(theme=theme) with open("output/article.md") as f: html = converter.convert(f.read())

6. Publish to draft box

6. 发布到草稿箱

api = WeChatAPI( appid=os.environ["WECHAT_APPID"], secret=os.environ["WECHAT_SECRET"] ) publisher = WeChatPublisher(api=api) media_id = api.upload_image(cover_path) draft_id = publisher.create_draft( title="选定标题", content=html, cover_media_id=media_id, author="作者名" ) print(f"✅ Published draft: {draft_id}")

---

api = WeChatAPI( appid=os.environ["WECHAT_APPID"], secret=os.environ["WECHAT_SECRET"] ) publisher = WeChatPublisher(api=api) media_id = api.upload_image(cover_path) draft_id = publisher.create_draft( title="选定标题", content=html, cover_media_id=media_id, author="作者名" ) print(f"✅ Published draft: {draft_id}")

---

References Claude Uses During Pipeline

Claude在流程中使用的参考文件

These files are read automatically by Claude when executing the skill:

File	Purpose
`references/topic-selection.md`	10-topic scoring rules (heat × fit × differentiation)
`references/frameworks.md`	5 article structure templates
`references/writing-guide.md`	Style rules, de-AI-ification techniques
`references/seo-rules.md`	WeChat SEO: title length, keyword density, tags
`references/visual-prompts.md`	Image generation prompt templates
`references/wechat-constraints.md`	WeChat HTML/CSS technical limits
`references/style-template.md`	Client style.yaml schema documentation

执行Skill时，Claude会自动读取以下文件：

文件	用途
`references/topic-selection.md`	10项选题评分规则（热度×契合度×差异化）
`references/frameworks.md`	5种文章结构模板
`references/writing-guide.md`	写作风格规则、去AI化技巧
`references/seo-rules.md`	微信公众号SEO规则：标题长度、关键词密度、标签
`references/visual-prompts.md`	图片生成提示词模板
`references/wechat-constraints.md`	微信HTML/CSS技术限制
`references/style-template.md`	客户style.yaml配置文件说明

Troubleshooting

故障排查

WeChat API 40001 / invalid credential

bash

undefined

微信API 40001 / 无效凭证

bash

undefined

Access token expires every 2 hours — the API wrapper auto-refreshes,

Access Token每2小时过期——API包装器会自动刷新，

but verify your appid/secret are correct:

但请确认你的appid/secret正确：

python3 -c " from toolkit.wechat_api import WeChatAPI import os api = WeChatAPI(os.environ['WECHAT_APPID'], os.environ['WECHAT_SECRET']) print(api.get_access_token()) "


**Image generation fails**
```bash

python3 -c " from toolkit.wechat_api import WeChatAPI import os api = WeChatAPI(os.environ['WECHAT_APPID'], os.environ['WECHAT_SECRET']) print(api.get_access_token()) "


**图片生成失败**
```bash

Test provider connectivity

测试服务商连通性

python3 toolkit/image_gen.py
--prompt "测试图片"
--output /tmp/test.png
--provider doubao

If doubao fails, switch to openai in config.yaml

如果doubao失败，在config.yaml中切换为openai


**Markdown conversion missing styles**
```bash


**Markdown转换丢失样式**
```bash

Verify theme loads correctly

验证主题是否正确加载

python3 -c "from toolkit.theme import load_theme; print(load_theme('tech-modern'))"

Preview output before publishing

发布前预览输出

python3 toolkit/cli.py preview article.md --theme tech-modern

Opens browser with rendered HTML

会在浏览器中打开渲染后的HTML


**Hotspot fetch returns empty**
```bash


**热点抓取返回空结果**
```bash

Platforms occasionally change their APIs — run with verbose:

平台偶尔会变更API——使用verbose模式运行：

python3 scripts/fetch_hotspots.py --limit 5 --verbose

Check which sources are failing; the script supports weibo/baidu/toutiao independently

检查哪个源失败；脚本支持单独使用weibo/baidu/toutiao


**Article not appearing in draft box**
- Ensure the WeChat account has **服务号** or **订阅号** API access enabled
- Check that your IP is whitelisted in the WeChat MP platform settings
- Draft box (`草稿箱`) requires `draft.add` API permission

---


**文章未出现在草稿箱**
- 确保微信账号已启用**服务号**或**订阅号**API权限
- 检查你的IP已在微信公众平台设置中加入白名单
- 草稿箱（`草稿箱`）需要`draft.add`API权限

---

Quick Reference

快速参考

bash

undefined

bash

undefined

Full standalone workflow

完整独立工作流

python3 scripts/fetch_hotspots.py --limit 20 --json > hotspots.json python3 scripts/seo_keywords.py --json "关键词1" "关键词2" python3 toolkit/image_gen.py --prompt "封面描述" --output cover.png --size cover python3 toolkit/cli.py preview article.md --theme tech-modern python3 toolkit/cli.py publish article.md --cover cover.png --title "标题"

Build playbook from historical articles

从历史文章构建规则手册

python3 scripts/build_playbook.py --client demo

Learn from edits

从编辑记录中学习

python3 scripts/learn_edits.py --original draft.md --edited final.md --client demo

Fetch article performance data

获取文章性能数据

python3 scripts/fetch_stats.py --article-id "msgid_here"

undefined

python3 scripts/fetch_stats.py --article-id "msgid_here"

undefined

wewrite-wechat-ai-publishing

Original

Translation

WeWrite — WeChat AI Publishing Skill

WeWrite — 微信公众号AI创作Skill

Installation

安装步骤

As a Claude Code Skill

作为Claude Code Skill安装

Clone the repo

Clone the repo

Or copy into an existing project

Or copy into an existing project

Python Dependencies

Python依赖安装

Configuration

配置设置

Triggering the Full Pipeline

触发全流程创作

Pipeline Steps & Scripts

流程步骤与脚本

1. Fetch Hotspots

1. 热点抓取

2. SEO Keyword Analysis

2. SEO关键词分析

3. Topic Selection

3. 选题生成

4. Framework & Writing

4. 框架与写作

5. AI Image Generation

5. AI图片生成

Cover image (recommended 900×383)

封面图（推荐尺寸900×383）

Inline content image

内文配图

6. Formatting — Markdown → WeChat HTML

6. 格式转换——Markdown转微信HTML

Preview in browser

在浏览器中预览

Available themes

查看可用主题

html is WeChat-ready with all inline styles

生成的html是符合微信要求的内联样式格式

7. Publish to WeChat Draft Box

7. 发布到微信草稿箱

Upload cover image first

先上传封面图

Push to draft box

推送至草稿箱

8. Fetch Article Stats (回填数据)

8. 获取文章数据（回填）

9. Learn from Manual Edits

9. 从人工编辑中学习

Client Configuration

客户配置

clients/my-tech-blog/style.yaml

clients/my-tech-blog/style.yaml

Edit style.yaml for your client

编辑style.yaml适配你的客户需求

Themes

主题样式

toolkit/themes/my-theme.yaml

toolkit/themes/my-theme.yaml

Full Pipeline — Python Orchestration

全流程Python编排

1. Fetch hotspots

1. 抓取热点

2. SEO analysis on top topics

2. 对热门话题进行SEO分析

3. (Claude selects topic, writes article — handled by SKILL.md)

3. (Claude负责选题和写作——由SKILL.md处理)

After Claude produces article.md:

当Claude生成article.md后：

4. Generate cover

4. 生成封面图

5. Convert Markdown → WeChat HTML

5. 转换Markdown为微信HTML

6. Publish to draft box

6. 发布到草稿箱

References Claude Uses During Pipeline