メニュー
Prepare and publish WeChat Official Account "贴图" image posts from a local image, title, and short caption. Use when the user asks to publish a 公众号贴图, 微信贴图, image-only WeChat post, or save one as a draft.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | wechat-image-post |
| description | Prepare and publish WeChat Official Account "贴图" image posts from a local image, title, and short caption. Use when the user asks to publish a 公众号贴图, 微信贴图, image-only WeChat post, or save one as a draft. |
Use this skill to turn a local image plus short copy into a WeChat Official Account "贴图" draft.
This skill supports:
drafts/<slug>/.This skill does not use wxp publish for true image posts. The existing article publisher creates article drafts through the official article draft workflow; the "贴图" editor in the WeChat backend is a separate web UI flow with fields for image, title, description, and optional location/reward/comments/links.
From the blog repository root:
node skills/wechat-image-post/scripts/prepare-wechat-image-post.mjs \
--image "C:/path/to/poster.png" \
--slug 20260625-codex-reconnecting-image-post \
--title "Codex 重连修复" \
--description "Codex 每次回答前都 Reconnecting？可能不是模型慢，而是 WebSocket 传输层不稳。改用户级 config，让 Responses API 直接走 HTTPS。"
Outputs:
drafts/<slug>/poster.pngdrafts/<slug>/caption.mddrafts/<slug>/wechat-image-post.jsondrafts/<slug>/wechat-image-post-publish.mdThe draft step depends on the web-access Chrome CDP proxy and an already logged-in WeChat Official Account browser session.
Before running:
node "$CLAUDE_SKILL_DIR/scripts/check-deps.mjs"
Then:
node skills/wechat-image-post/scripts/wechat-image-post-draft.mjs \
--manifest drafts/<slug>/wechat-image-post.json
The script opens the WeChat backend, enters the "贴图" editor, uploads the image, fills title and description, and clicks "保存为草稿".
If the script reports that CDP is unavailable, open Chrome remote debugging as instructed by web-access and rerun. If WeChat asks for login or QR confirmation, complete it in the browser, then rerun the command.
博客文章创作流水线。当用户提到写文章、新选题、继续某篇文章、审校、发布等创作相关任务时触发。自动检测当前文章处于哪个阶段，告知下一步该做什么，并扮演对应角色执行任务。即使用户只说"继续写那篇文章"或"帮我看看这篇"，也应该触发此 skill。
Download and extract public Douyin videos or image posts from user-provided douyin.com / v.douyin.com links. Use this skill whenever the user asks for 抖音下载, 抖音去水印, 抖音解析, 下载抖音视频, 保存抖音图集, or provides a Douyin video/share URL and wants local media files or media URLs.
Generate raster images (PNG/JPEG/WebP) using the user's ChatGPT subscription via a local one-file Python CLI — no OPENAI_API_KEY, no gateway, no daemon. Use when an agent needs to create a brand-new bitmap asset for the current project (photos, illustrations, icons, hero banners, mockups, sprites, concept art) and the output should be a bitmap file saved into the workspace. Do not use when the task is better solved by editing existing SVG/vector assets, writing code-native graphics (HTML/CSS/canvas), or extending an established repo icon system.