skilly. Buy ad slot
All skills
Documentation / AGENT SKILL

lark-doc

open.feishu.cn
721.6K installs
0

飞书云文档（Docx / Wiki）内容操作：读取、创建、编辑文档，插入或下载图片附件，以及操作思维笔记。用户提供文档 URL/token（包括 doubao.com 的 /docx/、/wiki/）时使用；按 URL 路径/token 而非域名路由。文档内嵌资源按读取参考中的统一规则分流。独立评论操作走 lark-drive；随正文读取评论使用 docs +fetch。表格或 Base 内部数据操作不在本 skill。

BEFORE YOU INSTALL

Understand the trade-offs.

SECURITY REVIEW

Finding details incomplete

An earlier scan reported medium risk. Some finding descriptions were not retained, so this result needs reassessment.

medium reported risk1 reported findingIncomplete evidence

1 finding description is unavailable. This record is eligible for a new scan; the earlier rating is retained for transparency.

Scope & how to read this review

Static inspection of the supplied skill source. “Complete” means the selected source and checks completed; it does not cover every downloaded package, external service, or runtime action.

Source coverage
partial
Reviewed
2026-09-24
Reviewed version
sha256:c6885
Latest catalog version
sha256:c6885

The reported findings score starts at 100 and subtracts 35 for each critical finding, 20 for high, 9 for medium, and 3 for low, with a floor of 0. It is shown only for complete, current reviews with all finding details. A score of 100 means no findings were reported in the reviewed source.

Severity describes potential impact, not the probability of harm. Review the conditions and evidence for each finding. A scan is not a safety guarantee.

SKILL QUALITY

Limited evidence for a verdict

35/100Provisional source score

The supplied SKILL.md provides clear routing for Lark document tasks and related capabilities. It functions primarily as an index: essential workflows and command details reside in references that were not supplied, so reliable end-to-end operation cannot be assessed.

low confidenceStatic reviewIntegration skill

The reviewer had limited evidence. These scores describe what was available, not a verified measure of reliability.

What works well

  • Clearly distinguishes reading, creation, import, editing, and related resource tasks.
  • Defines concrete local-path resolution rules and several important routing exceptions.
  • Provides an authentication-error recovery route and specific constraints for existing whiteboards and document copies.

Before you rely on it

  • Supply the referenced workflows and shared skill dependencies for assessment.
  • Document prerequisites, complete command examples, required inputs, and operation-specific recovery steps.
  • Add observable acceptance criteria for reading, creation, editing, media operations, and history restoration.
Clear instructionsCan your agent understand the task and expected outcome?3/4

Task boundaries and routing decisions are concrete, including exceptions for empty documents and unchanged imports. Detailed outcomes and execution contracts depend on unavailable references.

EVIDENCE
SKILL.md: "- **导入 / 空文档 — [`+create`](references/lark-doc-create.md)**：仅创建空文档或原样导入用户提供的完整内容时，跳过创建工作流。"

To improve: Specify the required inputs and expected returned result for each route, either here or in supplied references.

Complete workflowAre prerequisites, failure cases, and recovery covered?1/4

The file provides routing, identity guidance, and a limited recovery path, but lacks the prerequisites and ordered operational steps needed to complete its advertised tasks independently.

EVIDENCE
SKILL.md: "仅遇到未认证、token / 身份或 scope 错误时读取该 Skill，修复后重试。"

To improve: Supply the linked operational references, including CLI setup, input requirements, ordered steps, and recovery for failed or partially completed mutations.

Repeatable resultsAre examples and dependencies specific enough to repeat the work?1/4

Local resource lookup is deterministic, but complete command invocations, representative inputs, and expected outputs are absent from the supplied bundle.

EVIDENCE
SKILL.md: "XML 内的相对资源路径先查 CWD；仅文件不存在时再查源 XML 文件所在目录，同名文件以 CWD 为准。"

To improve: Provide reproducible command examples with sample document URLs or tokens, input content, and expected response fields; specify compatible CLI requirements where behavior depends on them.

Checkable outcomesCan you tell whether the task succeeded?1/4

A few observable constraints provide limited review criteria, but there are no supplied examples of successful results, post-operation checks, or tests. Listing a status command alone does not establish a verification procedure.

EVIDENCE
SKILL.md: "更新已有画板必须复用现有 token，禁止新建空白画板"

To improve: Define checks that confirm the target document, intended content changes, preserved resource identity, and final completion status for asynchronous operations.

Supporting filesDo the supplied scripts and references support the workflow?1/4

Supporting references are essential to this integration and are explicitly linked, but none were supplied. Their contents, completeness, and agreement with the routing instructions remain unknown.

EVIDENCE
SKILL.md: "- **从零创作 — [`创建工作流`](references/lark-doc-create-workflow.md)**：先完整执行创建工作流，**简单任务不是跳过的理由**；" Evidence from the referenced workflow was not available.

To improve: Include the referenced files and relevant cross-skill dependencies so their commands, workflow rules, and examples can be checked for coherence.

Review scope & scoring

Scores describe source evidence: 0 absent or contradicted, 1 weak, 2 incomplete, 3 solid, 4 strong. Supporting files are excluded when not applicable. Instruction skills can demonstrate quality through examples and checkable outcomes without executable tests.

Provisional source score
35/100
Files reviewed
1
Source coverage
skill only
Reviewed
2026-09-26
Rubric
2026-09-25-v3
Assessed version
sha256:c6885
Latest catalog version
sha256:c6885

Review limitations

  • Coverage is explicitly skill-only; referenced files and their consistency with SKILL.md could not be inspected.
  • This was a static review of supplied text only. No commands were executed and no runtime behavior was verified.

This review does not execute the skill or verify runtime results. Install counts and publisher reputation do not increase the score.

The full skill.

Original instructions from the publisher’s SKILL.md

# docs

## 场景与 Shortcut 路由

**CRITICAL：先判断场景，再读取该场景的参考文件；不要在任务开始时一次性读取全部参考文件。每个文件只在首次进入对应阶段时读取一次。**

**身份：文档操作推荐显式指定 `--as user`。**

**本地文件引用统一遵循文件访问权限：CWD 内优先使用 `@./相对路径`，其他目录使用 `@绝对路径`。XML 内的相对资源路径先查 CWD；仅文件不存在时再查源 XML 文件所在目录，同名文件以 CWD 为准。内联内容、stdin、在线文档没有源文件目录，不执行回退。**

### 文档内容

- **读取 / 摘要 — [`+fetch`](references/lark-doc-fetch.md)**：先读参考再获取文档。
- **从零创作 — [`创建工作流`](references/lark-doc-create-workflow.md)**：先完整执行创建工作流，**简单任务不是跳过的理由**；
- **导入 / 空文档 — [`+create`](references/lark-doc-create.md)**：仅创建空文档或原样导入用户提供的完整内容时，跳过创建工作流。
- **编辑 / block 直达链接 — [`+update`](references/lark-doc-update.md)**：语义改写、润色、重组、补写或排版均按 update 参考完成。

### 辅助能力

- **草稿初始化、解析与统计 — [`+script`](references/lark-doc-script.md)**：支持解析文档 URL / token 与本地 XML，统计字数并返回字符诊断；不支持 Markdown 输入。
- **历史版本 — [`+history-list` / `+history-revert` / `+history-revert-status`](references/lark-doc-history.md)**：查询、回滚文档历史版本或检查回滚任务状态。

### 资源、画板与思维笔记

- **插入本地素材 — [`+media-insert`](references/lark-doc-media-insert.md)**：在文末插入本地图片或文件。
- **预览素材 — [`+media-preview`](references/lark-doc-media-preview.md)**：预览文档或评论中的图片、附件或素材。
- **下载素材 — [`+media-download`](references/lark-doc-media-download.md)**：下载文档中的图片、附件、素材或画板缩略图。
- **Docx 封面 — [`+resource-download` / `+resource-update` / `+resource-delete`](references/lark-doc-resource-cover.md)**：下载、更新或删除 Docx 封面。
- **画板 — [`画板工作流`](references/lark-doc-whiteboard.md)**：创建或更新画板时先读取工作流；更新已有画板必须复用现有 token，禁止新建空白画板；使用 [`whiteboard +update`](../lark-whiteboard/references/lark-whiteboard-update.md) 写入。
- **思维笔记 — `mindnotes`**：已有思维笔记走 [`思维笔记链路`](references/lark-doc-mindnote.md)；新建思维笔记走 [`lark-doc-whiteboard`](references/lark-doc-whiteboard.md)。

### 认证与 Scope

执行 Shortcut 时，不预读 [`lark-shared`](../lark-shared/SKILL.md) 或预跑 `auth status --verify`；仅遇到未认证、token / 身份或 scope 错误时读取该 Skill，修复后重试。认证、身份或 scope 管理请求则直接使用该 Skill。

## 不在本 Skill 范围

- **Drive 文件级操作**：找文档、导入导出、云空间文件上传 / 下载 / 权限管理 → [`lark-drive`](../lark-drive/SKILL.md)。复制文档、创建副本或另存为副本时，按其指引使用 `lark-cli drive files copy`；不要用 `docs +fetch` + `docs +create` 重建正文。
- **独立评论操作**：添加、分页查看、回复评论或增删 reaction → [`lark-drive`](../lark-drive/SKILL.md)；只需紧凑评论上下文时，直接使用默认 JSON 响应的 `docs +fetch`。