메뉴
Use when a post from this repository should be appended to an existing Feishu doc or wiki page, especially when the user provides a target wiki/doc URL and wants update semantics with append as the default rather than creating a new child page.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when a post from this repository should be published as a new child page under the fixed Feishu wiki node `AI Coding 分享`, especially when the user says 上传到知识库、发到飞书知识库、新建一篇知识库文章、或要把某篇 content/posts 文章发到 AI Coding 分享。
Use when blog post has local images (img/ directory) that need to be visually understood, inserted into the article at appropriate positions, and/or migrated from local relative paths to a GitHub image hosting repository. Triggers: 插入图片, 配图, 上传图片到图床, 图片迁移, 图片外链, insert images, upload to image hosting.
Use when converting PDF files to markdown with embedded images extracted as separate files. Triggered by: "PDF转markdown", "PDF转md", "解析PDF", "提取PDF内容", "提取PDF图片", "convert PDF to markdown", "PDF to md", "extract PDF content".
将 Markdown 文件中的 mermaid 代码块替换为同目录 img/ 下对应的图片引用。 用户已手工将 mermaid 图导出为图片（png/gif/jpeg/webp/svg 等）放入 img/ 目录， 需要自动匹配并替换。 当 markdown 文件中同时存在 mermaid 代码块和 img/ 目录下的图片文件时使用。
从用户在对话中提供的文本创建 Hugo 博客文章，自动生成 front matter、运行 check-posts.py 检查、git commit 并 push。当用户贴一段文字并要求发博客、创建文章、发布文章、帮我发到博客、把这段文字变成博客、commit and push 这篇文章时使用。
| name | append-post-to-lark-doc |
| description | Use when a post from this repository should be appended to an existing Feishu doc or wiki page, especially when the user provides a target wiki/doc URL and wants update semantics with append as the default rather than creating a new child page. |
把本仓库里的文章追加到一个已经存在的飞书文档或 wiki 页面。
REQUIRED SUB-SKILL: Use $publishing-markdown-to-lark-docs for the actual upload workflow.
content/posts/.../index.md不要用于：
content/posts/ 下先定位唯一文章action: updateupdate_mode: appendoverwrite默认执行语义：
使用 $publishing-markdown-to-lark-docs 帮我把本地文章追加到已有飞书文档。
本次参数：
- input: <resolved post path>
- target: <user-provided doc/wiki target>
- action: update
- update_mode: append
先 dry-run，确认解析出的真实 doc token 和更新方式，再正式执行。每次写入后 fetch 回读确认。
完成后至少返回：
update_modecreateoverwrite