メニュー
Converts Markdown to WeChat Official Account HTML using AI-themed layouts and optionally one-click publishes to the draft box. Use when formatting Markdown for WeChat or uploading drafts with WECHAT_APPID/WECHAT_SECRET set in an env file.
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
Use when users ask in Chinese or English to manage local dev ports via natural language (for example: pick an available port, inspect who uses port 3000, free a port, run port conflict doctor, or initialize .portmanrc). Route to PortPilot CLI commands; run read-only actions automatically and require detailed confirmation for write actions.
Update or rewrite project README files directly with the current session model (Codex/Claude Code) using local repository facts, without calling external README generation APIs. Use when users ask to "update README", "rewrite readme.md", "generate project documentation", "refresh usage/install sections", or "align README with latest code changes" for a local repo.
Writes polished Chinese technical articles for WeChat or blog publication, then optionally generates matching illustrations. Use when user wants to write, draft, optimize, or illustrate a technical article. Handles the full pipeline: type selection → research → draft → fact-check → illustration.
Analyze any input material (URLs, code snippets, images, text, documents, screenshots) and extract technical insights, reusable code, and project ideas into the CodeSnippets knowledge base at /root/projects/CodeSnippets. Use this skill when the user drops a link, pastes code, shares a screenshot, sends a document, or provides raw text and wants it analyzed and archived as inspiration. Triggers: "收录", "分析这个", "记录一下", "add this to snippets", "collect", "/insight", or when user provides material (link/code/image/doc) with intent to archive into the project.
Create and edit JSON Canvas files (.canvas) with nodes, edges, groups, and connections. Use when working with .canvas files, creating visual canvases, mind maps, flowcharts, or when the user mentions Canvas files in Obsidian.
Create and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries. Use when working with .base files, creating database-like views of notes, or when the user mentions Bases, table views, card views, filters, or formulas in Obsidian.
| name | md2wechat |
| description | Converts Markdown to WeChat Official Account HTML using AI-themed layouts and optionally one-click publishes to the draft box. Use when formatting Markdown for WeChat or uploading drafts with WECHAT_APPID/WECHAT_SECRET set in an env file. |
Convert Markdown to WeChat Official Account HTML with inline CSS (AI mode) and optionally upload to the draft box.
# Render AI-style HTML (theme) from Markdown
python scripts/md_ai_render.py --md article.md --theme autumn-warm --out article.html
# Then publish
python scripts/wechat_publish.py --md article.md --html article.html --draft --cover cover.jpg
references/themes.md).python scripts/md_ai_render.py --md article.md --theme autumn-warm --out article.html
python scripts/wechat_publish.py --md article.md --html article.html --draft --cover cover.jpg
Use the first image as cover if the user does not provide one.
python scripts/wechat_publish.py --md article.md --html article.html --draft --cover cover.jpg --fetch-draft article.wechat.html
 in Markdown or ask in natural language.references/image-syntax.md.AI image generation is handled by Claude; no separate tool call.
Assume the user has an env file (e.g. .env) configured with:
WECHAT_APPIDWECHAT_SECRETOptional:
IMAGE_API_KEY (AI images)IMAGE_API_BASEThe publish tool loads .env by default; override with --env.
It also accepts AppID/AppSecret in the env file.
Draft title defaults to the Markdown filename (without .md). Override with --title.
Input (article.md):
# Hello World
This is a simple article.
- Item 1
- Item 2
Command:
python scripts/md_ai_render.py --md article.md --theme autumn-warm --out article.html
Output (article.html): WeChat-compatible HTML with inline CSS and autumn-warm theme styling.
Input: article.md + article.html + cover.jpg
Command:
python scripts/wechat_publish.py --md article.md --html article.html --draft --cover cover.jpg
Output:
{"success": true, "image_count": 2, "draft_media_id": "xxx"}
references/themes.md (AI themes and prompts)references/html-guide.md (WeChat HTML rules)references/image-syntax.md (image syntax)references/wechat-api.md (draft upload behavior)