← Back to Skills Marketplace
sh1-5

语雀知识库导出

by 拾伍 · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
209
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install yuque-export
Description
将语雀知识库的所有文档导出为本地 Markdown 文件,保留目录结构。需要 Chrome DevTools MCP。
Usage Guidance
This skill operates inside a connected Chrome DevTools MCP browser session and will run scripts and fetch requests using your logged-in Yuque session (cookies). That behavior is required to export private docs, but it means the skill has access to whatever the current browser profile can access. Before using: 1) only run with a browser/profile you trust and that is logged in to Yuque; 2) close or avoid having other sensitive sites open in that profile while the MCP session is used; 3) confirm the skill will only fetch yuque.com endpoints (SKILL.md indicates it targets Yuque APIs); and 4) if you have doubts, ask the publisher for the source or a code review — because this is instruction-only from an unknown source, exercise caution. If you want maximum safety, perform the export using a dedicated browser profile or session created solely for this task.
Capability Analysis
Type: OpenClaw Skill Name: yuque-export Version: 1.0.1 The skill is a legitimate utility designed to export Yuque knowledge bases to local Markdown files. It utilizes Chrome DevTools MCP to leverage the user's existing browser session for API access, injects a complex but transparent JavaScript parser (lakeToMarkdown) to convert Yuque's proprietary format, and uses standard Python/Bash commands to organize the directory structure and download images from Yuque's official CDN (cdn.nlark.com). All high-risk capabilities, such as local file writing and browser script execution, are strictly aligned with the stated purpose and lack any indicators of data exfiltration or malicious intent.
Capability Assessment
Purpose & Capability
The skill's name and description (export Yuque repo to Markdown) match the runtime instructions: it drives a browser DevTools session, discovers book_id and catalog_nodes, fetches docs, converts HTML to Markdown, and preserves directory structure. It does not request unrelated binaries, env vars, or credentials.
Instruction Scope
Runtime instructions require navigating the user’s browser session, listing network requests, injecting a large client-side conversion function, and issuing fetch() calls with credentials: 'include' so the browser session cookies are used. This is necessary for the stated task but is a high-privilege operation: MCP will see network activity and the skill runs scripts inside the logged-in browser context. The SKILL.md shows care (avoids /api/v2/repos/) and limits API targets to Yuque endpoints, but the actions could access other network logs or pages in the same browser session if misused.
Install Mechanism
Instruction-only skill with no install spec and no code files — nothing is downloaded or written by an installer. Lowest-risk install posture.
Credentials
No environment variables, credentials, or config paths are requested. The skill relies on an existing, logged-in browser session via Chrome DevTools MCP, which is proportional for exporting a user's Yuque content.
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges or to modify other skills. It runs ad-hoc against the connected MCP session and does not ask to persist tokens or enable itself permanently.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install yuque-export
  3. After installation, invoke the skill by name or use /yuque-export
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
yuque-export 1.0.1 - 新增支持自动保留语雀知识库的目录层级结构,导出时按分组/文件夹生成对应本地文件夹。 - 新增 `--images` 参数:可批量下载 Markdown 里的图片到本地 `_assets/` 目录,图片链接同时替换为本地相对路径。 - 目录和文档树改为通过 `/api/catalog_nodes` 获取,准确反映原有的文件夹和分组层级。 - 默认输出目录更新为 `~/Downloads/yuque-export`,可通过 `--output` 自定义,导出方式更灵活。 - lakeToMarkdown 解析功能增强,支持更多 card 卡片类型、公式、附件、复选框、外部/嵌入文档等。 - 优化说明文档,明确各参数功能和注意事项。
v1.0.0
Initial release of yuque-export: Export all documents from a Yuque knowledge base as local Markdown files. - Exports all documents from a specified Yuque knowledge base using a logged-in browser session (via Chrome DevTools MCP), bypassing the need for Yuque API tokens. - Automatically converts Yuque’s proprietary lake HTML format to Markdown, preserving code blocks and images. - Supports batch download with configurable output directory (default: ~/Downloads/yuque-export; customizable via --output). - Includes a Python script for splitting the exported combined Markdown file into individual files, one per document. - Implements rate-limited concurrent requests to avoid server throttling. - Handles empty documents, and notes that images are linked (not downloaded). - Guides users through each step from authentication, data extraction, conversion, to local download and splitting.
Metadata
Slug yuque-export
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is 语雀知识库导出?

将语雀知识库的所有文档导出为本地 Markdown 文件,保留目录结构。需要 Chrome DevTools MCP。 It is an AI Agent Skill for Claude Code / OpenClaw, with 209 downloads so far.

How do I install 语雀知识库导出?

Run "/install yuque-export" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 语雀知识库导出 free?

Yes, 语雀知识库导出 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 语雀知识库导出 support?

语雀知识库导出 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 语雀知识库导出?

It is built and maintained by 拾伍 (@sh1-5); the current version is v1.0.1.

💬 Comments