lark-workflow-doc-to-slides
Skill2 starsChanged 8 months agoApache-2.0View on GitHub
文档转幻灯片工作流:把飞书文档或 Wiki 内容先整理成可审阅的 slide outline,经用户确认后新建 Slides 或追加到已有 Slides。适用于使用 doc_url、doc_token、doc_name 作为来源,或把文档内容追加到 target_slides_url 指向的现有演示文稿。
In this repository27 other files it gives its agents.
- ai-hotspot-collectorSkill
- article-multi-publisherSkill
- article-pipeline-skillSkill
- character-ip-risk-checkSkill
- chrome-extension-devSkill
- chrome-extension-e2e-automationSkill
- chrome-extension-publishSkill
- chrome-extension-social-promoSkill
- chrome-webstore-image-generatorSkill
- code-reviewSkill
- fund-querySkill
- github-trending-writerSkill
- notionSkill
- playwright-twitter-accessSkill
- post-to-xhsSkill
- rewrite-skillSkill
- small-capital-investingSkill
- technical-article-preflightSkill
- technical-article-reviewSkill
- technical-spoken-rewriteSkill
- tencent-dev-community-publisherSkill
- toutiao-publisherSkill
- url-readerSkill
- wechat-article-formatterSkill
- wechat-article-generatorSkill
- wechat-publisherSkill
- wechat-sticker-publisherSkill
The file
Files
Lark Workflow Doc To Slides
把文档变成 PPT 时,先把“文档内容”变成“可审阅的大纲”,再把已批准的大纲变成 Slides。
适用触发语义:
- “把这篇文档转成飞书幻灯片 / PPT”
- “根据
doc_url/doc_token/doc_name生成 Slides” - “把这个 Wiki 做成汇报幻灯片”
- “把这篇文档追加到现有 Slides”
- “用
target_slides_url继续往当前 deck 里加几页”
CRITICAL — 开始前必须先读取已安装的 lark-shared、lark-doc、lark-slides skills;本机默认路径是 ~/.codex/skills/lark-shared/SKILL.md、~/.codex/skills/lark-doc/SKILL.md、~/.codex/skills/lark-slides/SKILL.md。
Required Inputs
必须且只能提供一个来源标识:
doc_urldoc_tokendoc_name
可选输入:
target_slides_urlcontent_mode:faithful|reporttitlemax_slides
默认值:
- 未提供
target_slides_url时,target_mode = new - 提供了
target_slides_url时,target_mode = append - 未提供
content_mode时,默认report
Workflow Invariants
- 必须先生成 outline,再做 XML render 和 Slides 发布。
- 用户未明确确认 outline 前,禁止创建新 deck,也禁止向已有 deck 追加页面。
doc_name命中多个候选时,必须停下来让用户选;禁止自动猜测。- 搜索结果里的 wiki 命中只能当候选来源;在 fetch 前,必须先按已安装的
lark-doc指南解析成底层对象类型 / token,不能把 wiki 搜索命中当成“已可直接抓取”的来源。 - append 模式只允许新增 slides;禁止重写、删除、重排已有页面。
- append 模式默认不生成通用“封面 / 目录”;只有用户明确要求新章节分隔页时,才允许带
section_divider: true的 divider 页。
Routing
- 无
target_slides_url:- 读取
references/workflow-new-slides.md - 再读取
references/content-modes.md - 再读取
references/slide-authoring-rules.md
- 读取
- 有
target_slides_url:- 读取
references/workflow-append-slides.md - 再读取
references/content-modes.md - 再读取
references/slide-authoring-rules.md
- 读取
Execution Shape
执行脚本位于 scripts/doc_to_slides.py。工作流子命令:
python3 lark-workflow-doc-to-slides/scripts/doc_to_slides.py resolve-source ...
python3 lark-workflow-doc-to-slides/scripts/doc_to_slides.py fetch ...
python3 lark-workflow-doc-to-slides/scripts/doc_to_slides.py validate-outline ...
python3 lark-workflow-doc-to-slides/scripts/doc_to_slides.py render ...
python3 lark-workflow-doc-to-slides/scripts/doc_to_slides.py publish ...运行目录约定:
.lark-workflow-doc-to-slides/runs/<timestamp>-<slug>/常见产物:
resolved-source.jsonsource.jsonsource.mdoutline.jsonslides.jsonrender-summary.jsonpublish-result.json
publish-result.json 采用归一化的顶层字段:
target_modexml_presentation_idurlslide_idsslides_addedrun_dir
发布保护:
publish只接受与当前outline.json匹配的slides.json- 如果
render-summary.json缺失,或outline/slides指纹不一致,必须先重新 render
Operator Rules
- 用 AI 负责理解原文和写 outline;用脚本负责校验、渲染、发布。
- 生成 outline 时,优先对齐
templates/outline.json的字段形状;每页 slide 保留显式布尔字段section_divider,默认false,append 模式需要章节分隔页时才设为true。 - 发布阶段只允许使用现有
lark-cli命令链;不存在slides +create-from-outline这种快捷命令。 - 如果权限或身份不对,按
lark-shared规则先修正身份与授权,再继续工作流。 - 涉及 wiki source / wiki target 时,按当前已安装 refs / schema 查看真实返回字段,不要在 skill 文案里猜测自定义响应包裹层。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

