wechat-article-pipeline
starsforks
How to install wechat-article-pipeline?
jhwreal/wechat-article-pipeline-skill/wechat-article-pipeline
$npx -y skills add jhwreal/wechat-article-pipeline-skill --skill wechat-article-pipelineUse when a user wants a publish-ready Chinese WeChat/公众号 article package, editable article HTML workbench, article formatting, image-job planning, missing visual jobs, official WeChat draft-box delivery, synchronized 今日头条 draft/publishing through Chrome, or explicitly says "打开秘书模
SKILL.md
| 1 | # WeChat Article Pipeline |
| 2 | |
| 3 | Produce a complete local article package from the user's topic, draft, notes, or markdown. Call delivery stages only when requested. |
| 4 | |
| 5 | ## Core Decisions |
| 6 | |
| 7 | - If the user asks to write, make, package, format, or polish a WeChat/公众号 article, proceed with this skill. |
| 8 | - If the user says "打开秘书模式", enable it for this request only and read its section in [style-guide.md](references/style-guide.md). Do not infer or mention it unless asked. |
| 9 | - If the user gives only a rough idea, infer the brief and write. Ask only when ambiguity would change the article conclusion, audience, legal risk, account selection, or delivery target. |
| 10 | - If the user asks for "不配图", "只排版", "直接格式化", or similar, use the no-image path. |
| 11 | - If the user asks to补图, continue, or fix missing assets, use the missing-image path and do not rebuild finished images. |
| 12 | - If the user asks to导入草稿箱, create a WeChat draft through official APIs only. Never use browser automation or private `mp.weixin.qq.com` endpoints for delivery. |
| 13 | - For explicitly requested Toutiao sync or publishing, use Chrome and read [publishing-toutiao.md](references/publishing-toutiao.md). This does not change the WeChat API-only rule. |
| 14 | |
| 15 | ## Workspace Contract |
| 16 | |
| 17 | Keep final artifacts in the current workspace unless the user names another location: |
| 18 | |
| 19 | - markdown: `<workspace>/files/<slug>.md` |
| 20 | - focused markdown: `<workspace>/files/<slug>.focused.md` |
| 21 | - image jobs: `<workspace>/files/<slug>.image-jobs.json` |
| 22 | - HTML workbench: `<workspace>/files/<slug>.html` |
| 23 | - job: `<workspace>/files/<slug>.job.json` |
| 24 | - optional publish manifest: `<workspace>/files/<slug>.publish-manifest.json` |
| 25 | - images: `<workspace>/image/<slug>/cover.png`, `body-*.png`, `closing.png` |
| 26 | |
| 27 | Keep final assets out of temp directories and `$CODEX_HOME/generated_images`. |
| 28 | |
| 29 | ## Default Article Path |
| 30 | |
| 31 | 1. Inspect `files/` and `image/` before choosing a slug. |
| 32 | 2. Draft in markdown first. Use [workflow.md](references/workflow.md) and [style-guide.md](references/style-guide.md). |
| 33 | 3. Place visual placeholders in the markdown only when images are desired: `cover`, `body-1`, `body-2`, ..., `closing`. |
| 34 | 4. Run the orchestration script once for planning: |
| 35 | |
| 36 | ```bash |
| 37 | python3 <skill>/scripts/postprocess_wechat_article.py \ |
| 38 | <workspace>/files/<slug>.md \ |
| 39 | <workspace>/files/<slug>.html \ |
| 40 | --workspace <workspace> \ |
| 41 | --article-slug <slug> \ |
| 42 | --jobs-out <workspace>/files/<slug>.image-jobs.json \ |
| 43 | --focused-article-out <workspace>/files/<slug>.focused.md \ |
| 44 | --support-dir <workspace>/files/wechat-article-pipeline/<slug> \ |
| 45 | --plan-only |
| 46 | ``` |
| 47 | |
| 48 | 5. Read [image-production.md](references/image-production.md), then generate `generation_queue[]` once. Use at most `min(queue length, currently available worker slots)` workers, refilling freed slots. Save each result to its `output`. Do not inspect generated images; continue to packaging and verification. |
| 49 | 6. Run the same script again without `--plan-only` to build the HTML, job JSON, crop previews, and support files. Add `--publish-manifest` only when the user requested API draft-box handoff. |
| 50 | 7. Run `verify_wechat_article_package.py <workspace>/files/<slug>.html` and fix any failures before delivery. |
| 51 | 8. After every verified HTML workbench build, start the local persistence server before delivery. This is the default for any task that produces an HTML workbench, even when the user did not explicitly ask to open it: |
| 52 | |
| 53 | ```bash |
| 54 | python3 <skill>/scripts/serve_wechat_workbench.py \ |
| 55 | <workspace>/files/<slug>.html \ |
| 56 | --workspace <workspace> |
| 57 | ``` |
| 58 | |
| 59 | Keep it running. Deliver `WORKBENCH_URL` first and `HTML_PATH` second. Direct-file mode is preview/copy-only; use the loopback URL for editing and lock editing if persistence disconnects. |
| 60 | |
| 61 | ## Fast Paths |
| 62 | |
| 63 | No-image formatting: |
| 64 | |
| 65 | ```bash |
| 66 | python3 <skill>/scripts/postprocess_wechat_article.py \ |
| 67 | <workspace>/files/<slug>.md \ |
| 68 | <workspace>/files/<slug>.html \ |
| 69 | --no-images \ |
| 70 | --support-dir <workspace>/files/wechat-article-pipeline/<slug> |
| 71 | ``` |
| 72 | |
| 73 | Use this when the user wants a formatted article without generated visuals. Existing `{{visual:*}}` image placeholders are removed from the packaged copy, so an illustrated source can safely use this path without manual cleanup. This path skips publish-manifest creation by default; add `--publish-manifest` only for API handoff. Draft creation may still need a cover asset; state that clearly. |
| 74 | |
| 75 | No body images, but draft-box delivery: |
| 76 | |
| 77 | ```bash |
| 78 | python3 <skill>/scripts/postprocess_wechat_article.py \ |
| 79 | <workspace>/files/<slug>.md \ |
| 80 | <workspace>/files/<slug>.html \ |
| 81 | --no-images \ |
| 82 | --publish-manifest \ |
| 83 | --cover-image <workspace>/image/<slug>/cover.png |
| 84 | ``` |
| 85 | |
| 86 | Use this w |
People who liked this also viewed
6 picksType
- azure-resource-lookupList, find, and show Azure resources across subscriptions or resource groups. Handles prompts like \"list the websites in my subscription\", \"list my web apps\", \"show my app services\", \"list virtSkillsJul 2026
- azure-messagingTroubleshoot and resolve issues with Azure Messaging SDKs for Event Hubs and Service Bus. Covers connection failures, authentication errors, message processing issues, and SDK configuration problems. SkillsJul 2026
- lark-doc飞书云文档（Docx / Wiki 文档）：读取和编辑飞书文档内容。当用户给出文档 URL 或 token，或需要查看、创建、编辑文档、插入或下载文档图片附件时使用。文档中嵌入的电子表格、多维表格、画板，先用本 skill 提取 token 再切到对应 skill。当用户给出 doubao.com 的 /docx/ 或 /wiki/ URL/token 时，也应直接使用本 skill；路由依据是 SkillsJul 2026
- lark-im飞书即时通讯：收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件（支持大文件分片下载）、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片（Interactive Card）、监听卡片按钮回调（card.action.trigger）。当用户需要发消息、查看或搜索聊天记录、下载聊天中的文件、查看群成员、搜索群、创建群聊或话题群、管理标记数据、管理 FeedSkillsJul 2026
- lark-wiki飞书知识库：管理知识空间、空间成员和文档节点。创建和查询知识空间、查看和管理空间成员、管理节点层级结构、在知识库中组织文档和快捷方式。当用户需要在知识库中查找或创建文档、浏览知识空间结构、查看或管理空间成员、移动或复制节点时使用。当用户给出 doubao.com 的 /wiki/ URL/token 时，也应直接使用本 skill，不要因为域名不是飞书而回退到 WebFetch；路由依据是 URLSkillsJul 2026
- lark-mail飞书邮箱：Use when user mentions 起草邮件、写邮件、草稿、发送/回复/转发邮件、查阅邮件、看邮件、搜索邮件、邮件文件夹、邮件标签、邮件联系人、监听新邮件、邮件收信规则等；use for mail/email intent only. Do not use for docs/sheets/calendar/auth setup/pure contact lookup/IM chSkillsJul 2026
