Lark Doc
Let your coding agent create, read, summarize, and edit Feishu (Lark) cloud documents via `lark-cli` v2 when specs and docs live in your team wiki.
Overview
lark-doc is an agent skill most often used in Build (also Validate and Grow) that reads, creates, and updates Feishu cloud documents through `lark-cli` docs v2 commands.
Install
npx skills add https://github.com/larksuite/cli --skill lark-docWhat is this skill?
- Fetch, summarize, and partially read Feishu docs with v2 `docs +fetch` scopes and structured excerpt output
- Create and patch Docx content with `--api-version v2` using DocxXML (or Markdown on request) via `docs +create` and `doc
- Hands off embedded sheets, Bitable, and whiteboard assets to sibling Lark skills after extracting tokens from the docume
Adoption & trust: 211k installs on skills.sh; 13.7k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
Your specs and notes are locked in Feishu Docx URLs and your agent cannot reliably fetch, summarize, or patch that content without the correct v2 CLI flags and DocxXML rules.
Who is it for?
Solo builders and tiny teams on Feishu who want agents to open doc links, extract content for planning or coding, and write changes back with DocxXML or Markdown.
Skip if: Teams not on Lark/Feishu, or workflows that only need spreadsheets, calendars, or chat without touching cloud documents.
When should I use this skill?
User provides a Feishu/Lark document URL or token, or asks to view, read, open, summarize, create, translate, review, or edit cloud Docx/wiki content—including embedded sheets, Bitable, whiteboards, or sync blocks where
What do I get? / Deliverables
You get v2-backed fetch excerpts or updated Lark documents with append, replace, or structural edits aligned to Feishu Docx block semantics.
- Structured document excerpts or full fetch output from `docs +fetch --api-version v2`
- New or updated Feishu Docx pages via create/update commands
- Extracted tokens for handoff to Lark sheet, Bitable, or whiteboard skills
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Product and agent-readable documentation usually lives in the build phase; Feishu Docx is a common home for specs, runbooks, and shipped feature notes for solo builders on Lark. The skill is narrowly focused on cloud document CRUD and content extraction, which maps to the docs subphase rather than generic backend or PM tooling.
Where it fits
Fetch a PRD in Feishu and extract excerpts so the agent can turn scope bullets into a build checklist.
Create a new architecture doc with `docs +create --api-version v2` and append API sections after each shipping milestone.
Read meeting notes from a shared Lark doc and rewrite action items into a structured task section.
Pull an existing help-center wiki page, translate or refresh copy, and push updates without leaving the agent session.
How it compares
A Lark Docx CLI integration skill—not a generic markdown file editor or an MCP document server.
Common Questions / FAQ
Who is lark-doc for?
lark-doc is for solo and indie builders who standardize on Feishu for specs, wikis, and operational docs and want their coding agent to operate those docs via `lark-cli` v2.
When should I use lark-doc?
Use it when you have a Feishu document URL or token and need to view, summarize, translate, create, or edit Docx content; during Validate for scope docs, Build for technical documentation, or Grow when refreshing customer-facing or internal knowledge base pages.
Is lark-doc safe to install?
It requires the `lark-cli` binary and network access to your Lark tenant; review the Security Audits panel on this Prism page and confirm OAuth scopes in lark-shared before letting an agent update production wikis.
SKILL.md
READMESKILL.md - Lark Doc
# docs (v2) > **⚠️ API 版本：本 skill 使用 v2 API。所有 `docs +create --api-version v2`、`docs +fetch --api-version v2`、`docs +update --api-version v2` 命令必须携带 `--api-version v2`。** ```bash # 常用示例 lark-cli docs +fetch --api-version v2 --doc "文档URL或token" lark-cli docs +create --api-version v2 --content '<title>标题</title><p>内容</p>' lark-cli docs +update --api-version v2 --doc "文档URL或token" --command append --content '<p>内容</p>' ``` ## 前置条件 — 执行操作前必读 **CRITICAL — 执行对应操作前，MUST 先用 Read 工具读取以下文件，缺一不可：** 1. [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md) — 认证、权限处理、全局参数（所有操作通用） 2. **读取文档（`docs +fetch --api-version v2`）** → 必读 [`lark-doc-fetch.md`](references/lark-doc-fetch.md)（`--scope` / `--detail` 选择、局部读取策略、`<fragment>` / `<excerpt>` 输出结构） 3. **创建或编辑文档内容** → 必读 [`lark-doc-xml.md`](references/lark-doc-xml.md)（XML 语法规则，仅当用户明确要求 Markdown 时改读 [`lark-doc-md.md`](references/lark-doc-md.md)）；从零创建时加读 [`lark-doc-create-workflow.md`](references/style/lark-doc-create-workflow.md)；编辑已有文档时加读 [`lark-doc-update-workflow.md`](references/style/lark-doc-update-workflow.md) **未读完以上文件就执行相应操作会导致参数选择错误、格式错误或样式不达标。** > **格式选择规则（全局）：** > - **创建 / 导入场景**（`docs +create`，或 `docs +update --command append/overwrite` 的整段写入）：XML 和 Markdown 都可以。用户提供 `.md` 本地文件、或明确说"导入 Markdown"时，直接用 Markdown；否则默认 XML（可用 callout、grid、checkbox 等富 block）。 > - **精准编辑场景**（`docs +update` 的 `str_replace` / `block_insert_after` / `block_replace` / `block_delete` / `block_move_after` 等局部精修指令）：优先使用 XML（`--doc-format xml`，即默认值）。XML 能稳定表达 block 结构和样式，局部精修更可控；不要因为 Markdown 更简单就自行切换。 ## 快速决策 - 用户需要在文档内**创建、复制或移动**资源块（画板、电子表格、多维表格等）时，必须先读取 [`lark-doc-xml.md`](references/lark-doc-xml.md) 的「三、资源块」章节 - 写文档时，重要信息（核心流程、架构、对比、风险、路线图、关键指标、因果关系）优先规划为画板，不要只用文字或表格承载 - 新增画板必须隔离到 SubAgent：简单图由 SubAgent 直接插入 `<whiteboard type="svg">完整 SVG</whiteboard>`，不读 `lark-whiteboard`；复杂图才由主 Agent 先建 `<whiteboard type="blank"></whiteboard>`，再启动 SubAgent 读取 `lark-whiteboard` 写入 - 用户说"看一下文档里的图片/附件/素材""预览素材" → 用 `lark-cli docs +media-preview` - 用户明确说"下载素材" → 用 `lark-cli docs +media-download` - 如果目标是画板/whiteboard/画板缩略图 → 只能用 `lark-cli docs +media-download --type whiteboard`（不要用 `+media-preview`） - 用户说"找一个表格""按名称搜电子表格""找报表""最近打开的表格""最近我编辑过的 xxx" → 直接用 `lark-cli drive +search`（参考 [`lark-drive`](../lark-drive/references/lark-drive-search.md)）。**老的 `docs +search` 已进入维护期、后续会下线，不要再新增依赖。** - `drive +search` 结果里会直接返回 `SHEET` / `Base` / `FOLDER` 等云空间对象，是资源发现的统一入口 - 拿到 spreadsheet URL/token 后 → 切到 `lark-sheets` 做对象内部操作 - 用户说"给文档加评论""查看评论""回复评论""给评论加/删除表情 reaction" → 切到 `lark-drive` 处理 - 文档内容中出现嵌入的 `<sheet>`、`<bitable>` 或 `<cite file-type="sheets|bitable">` 标签时 → **必须主动提取 token 并切到对应技能下钻读取内部数据**，不能只呈现标签本身 | 标签 / 属性 | 提取字段 | 切到技能 | |-|-|-| | `<sheet token="..." sheet-id="...">` | `token` -> spreadsheet_token, `sheet-id` | [`lark-sheets`](../lark-sheets/SKILL.md) | | `<bitable token="..." table-id="...">` | `token` -> app_token, `table-id` | [`lark-base`](../lark-base/SKILL.md) | | `<cite type="doc" file-type="sheets" token="..." sheet-id="...">` | 同 `<sheet>` | [`lark-sheets`](../lark-sheets/SKILL.md) | | `<cite type="doc" file-type="bitable" token="..." table-id="...">` | 同 `<bitable>` | [`lark-base`](../lark-base/SKILL.md) | | `<synced_reference src-token="..." src-block-id="...">` | `src-token` -> doc_token, `src-block-id` -> block_id | 用 `docs +fetch --api-version v2` 读取 src-token 文档，定位 block | **补充：** 云空间资