公众号贴图|九宫格|多图推送|图片消息|表情包 — 贴图与多图推送:从创意构思、AI 生图到公众号图片消息发布全流程,含九宫格自动排布、多图压缩、组图打包。面向公众号运营、自媒体、IP 账号。触发词:「贴图」「多图推送」「发组图」「图片消息」「九宫格」「做一组图」「图片帖子」「发几张图」「不写正文发图」「只发图不写字」。是文章内配图/封面请走 aws-wechat-article-images;需要多环节串联(写+审+排+配图+发)请走 aws-wechat-article-main。
npx skills add https://github.com/aiworkskills/wechat-article-skills --skill aws-wechat-sticker
贴图 / 多图推送一条龙 —— 构思→生图→图片消息群发,九宫格自动排布、多图打包。
> 套件说明 · 本 skill 属 aws-wechat-article-* 一条龙套件(共 9 个 slug,入口 aws-wechat-article-main)。跨 skill 的相对引用依赖同一 skills/ 目录,建议一并 clawhub install 全套。源码:<https://github.com/aiworkskills/wechat-article-skills>
本 skill 编排贴图与多图推送,调用同套件的 image_create.py(images)生图;可选调用 publish.py 发布。会把图片 prompt 发给外部图像 API;若走发布路径,会把图片作为 POST body 上传到微信 API。 具体行为:
aws.env 的 IMAGE_MODEL_API_KEY;走发布时额外读取 WECHAT_{N}_APPID / WECHAT_{N}_APPSECRETIMAGE_MODEL_API_KEY 以 Authorization: Bearer 头发送到 image_model.base_url 外部端点;WECHAT_{N}_APPID/APPSECRET 用于换 access_token 后调用微信 APImaterial/add_material / draft/add.aws-article/config.yaml、本篇 article.yaml、imgs/prompts/*.mdimgs/*.{png,webp}、imgs/outline.md、article.yaml 状态字段python3 {baseDir}/../aws-wechat-article-images/scripts/image_create.py;可选 python3 {baseDir}/../aws-wechat-article-publish/scripts/publish.py本 skill 属 aws-wechat-article-* 一条龙公众号套件的贴图/多图推送专用分支(入口 aws-wechat-article-main)。工作流中的生图步骤与发布步骤依赖同套件其他 slug 提供的能力。
aws-wechat-article-images(用于生图);若要走发布链路,建议再装 aws-wechat-article-publish。skills/ 根目录时,跨 skill 引用都能读到。完整 9 slug 清单见 源码仓库。
长文图文(标题+正文+插图+后台发文)→ aws-wechat-article-main;长文内单篇插图 → aws-wechat-article-images。
创作以图片为主的公众号内容:多张图片 + 每张配文,统一风格。
imgs/(outline + prompts + 图片)与可发布的多图内容任何操作执行前,必须按 首次引导 执行其中的 「检测顺序」。检测通过后才能进行以下操作(或用户明确书面确认「本次不检查」):
从选题到发布的前置规则见 aws-wechat-article-main/SKILL.md;本 skill 描述贴图子流程。
图片模型:image_model(provider、base_url、model 等)在 config.yaml;IMAGE_MODEL_API_KEY 在 aws.env。键名对照 {baseDir}/../aws-wechat-article-main/references/env.example.yaml(与 image_create.py 一致)。
交互约定:须遵守 main 的智能体行为约束——未通过环境校验且未获用户明确「本次例外」时,不得假装已走专用生图 API。
Agent 执行:确定本 SKILL.md 所在目录为 {baseDir}。
| 脚本 | 路径(相对仓库 skills/) | 用途 |
|------|---------------------------|------|
| image_create.py | aws-wechat-article-images/scripts/image_create.py | 专用生图 API:读取 .aws-article/config.yaml 的 image_model + 根目录 aws.env 的 IMAGE_MODEL_API_KEY |
| publish.py | {baseDir}/../aws-wechat-article-publish/scripts/publish.py | 发布(与 publish skill 一致) |
贴图进度:
- [ ] 第1步:环境检查 + 本篇约束(合并 YAML)
- [ ] 第2步:确定选题
- [ ] 第3步:确定风格
- [ ] 第4步:规划图序
- [ ] 第5步:展示方案并等待确认 ⛔
- [ ] 第6步:生成图片(**脚本失败时**见同节「调用失败」分支)
- [ ] 第7步:审稿
- [ ] 第8步:发布
若本篇在 drafts/…/ 下:按 .aws-article/config.yaml → 本篇 article.yaml 合并读取约束(同键本篇优先,最后层覆盖),重点字段:multi_image_count、tone、target_reader、custom_sticker_style > default_sticker_style(须为 YAML 字符串列表;custom_* 非空时优先于 default_*;多候选时智能体择一并写回本篇为单元素列表)。若无有效 YAML,以用户口述主题 + skill 默认值作为约束。
topics 产出的贴图卡片 / 用户直接给主题 / 用户提供素材图片。
贴图风格加载优先级:
custom_sticker_style > default_sticker_style(若有;custom_* 优先;列表多元素时须先择一).aws-article/presets/sticker-styles/ 下的自定义风格全组图统一风格。
产出 imgs/outline.md:每张图的用途、文案要点、文件名。详见:references/workflow.md
生成方式(优先级):
image_create.py — 依赖 config.yaml 的 image_model + aws.env 的 IMAGE_MODEL_API_KEY必须告知用户当前使用的方式:
ℹ️ 使用 image_create.py 调用专用生图模型({model})ℹ️ 本次未走 image_create.py(原因:…)调用专用 API 时(在仓库根执行,路径按本篇 imgs/ 调整):
python {baseDir}/../aws-wechat-article-images/scripts/image_create.py batch drafts/YYYYMMDD-slug/imgs/prompts/ -o drafts/YYYYMMDD-slug/imgs/
单张:python {baseDir}/../aws-wechat-article-images/scripts/image_create.py generate imgs/prompts/01.md -o imgs/01.png
连通性自检:python {baseDir}/../aws-wechat-article-images/scripts/image_create.py test
图片内文字与 prompt 构建规则与长文配图一致,见 aws-wechat-article-images/SKILL.md 与 prompt-construction.md。
image_create.py 调用失败时(智能体必选分支)沿用配图 skill 的同一规则:必须摘要 stderr 关键行,按 网络 / 配置·凭证 / 业务·参数 分类处理;禁止配置错误时静默降级。细则参照 aws-wechat-article-images/SKILL.md 第 6 步「调用失败」表格。
贴图专用清单:references/checklist.md
python {baseDir}/../aws-wechat-article-publish/scripts/publish.py full article/
| 读取 | 产出 |
|------|------|
| topic-card.md(可选)、.aws-article/config.yaml + 本篇 article.yaml | imgs/(outline + prompts + 图片) |
Analyze text and images for harmful content with customizable blocklists.
Azure AI Content Understanding SDK for Python. Use for multimodal content extraction from documents, images, audio, and video.
Azure OpenAI SDK for .NET. Client library for Azure OpenAI and OpenAI services. Use for chat completions, embeddings, image generation, audio transcription, and assistants.
Azure AI Voice Live SDK for .NET. Build real-time voice AI applications with bidirectional WebSocket communication.
Build real-time voice AI applications with bidirectional WebSocket communication.
Azure Speech to Text REST API for short audio (Python). Use for simple speech recognition of audio files up to 60 seconds without the Speech SDK.
Expert at handling file uploads and cloud storage. Covers S3, Cloudflare R2, presigned URLs, multipart uploads, and image optimization. Knows how to handle large files without blocking.
Expert Mermaid diagram creation, validation, and rendering with dual-engine output (SVG/PNG/ASCII). Supports all 20+ diagram types including C4 architecture, AWS architecture-beta with service icons, flowcharts, sequence, ERD, state, class, mindmap, timeline, git graph, sankey, and more. Features code-to-diagram analysis, batch rendering, 15+ themes, and syntax validation. Use when users ask to create diagrams, visualize architecture, render mermaid files, generate ASCII diagrams, document system flows, model databases, draw AWS infrastructure, analyze code structure, or anything involving "mermaid", "diagram", "flowchart", "architecture diagram", "sequence diagram", "ERD", "C4", "ASCII diagram". Do NOT use for non-Mermaid image generation, data plotting with chart libraries, or general documentation writing.
Take aiworkskills/aws-wechat-sticker from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.