← 返回 Skills 市场
367
总下载
0
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install notebooklm-audio-generator
功能描述
Automates uploading multiple sources (files, URLs, YouTube, Drive, text) to a NotebookLM notebook, generating a deep dive audio overview in a preferred langu...
安全使用建议
This skill roughly does what it says, but exercise caution before running it: 1) Verify the notebooklm-mcp-cli package is the official client (check PyPI/GitHub) rather than installing an untrusted package. 2) Do not blindly run the suggested 'git clone' + './run.sh' from an unfamiliar GitHub repo; inspect the repo and run.sh first (or prefer a vetted epub conversion tool). 3) Understand that 'nlm login' and Drive document uploads will access your Google account/data; only proceed if you trust the NotebookLM client and its consent screens. 4) If you want to install this skill, ask the author for authoritative sources (official CLI docs, secure install instructions) or replace the epub step with a verified converter. 5) If you are not comfortable inspecting/run arbitrary scripts, do not grant this skill file-system or shell execution privileges.
功能分析
Type: OpenClaw Skill
Name: notebooklm-audio-generator
Version: 1.0.0
The skill in SKILL.md utilizes high-risk capabilities including the execution of shell commands and a local script ('run.sh') cloned from an external GitHub repository (github.com/SPACESODA/epub2txt.git). It also employs OS-level GUI pickers (e.g., osascript, zenity) to access the file system. While these actions are plausibly necessary for the stated purpose of processing files for NotebookLM, the reliance on unverified external code and direct shell execution warrants a suspicious classification.
能力评估
Purpose & Capability
The name/description (create NotebookLM audio overviews) aligns with the actions described (use notebooklm-mcp-cli, upload files/URLs/Drive, generate audio). Requiring epub conversion and a NotebookLM CLI makes sense. However, the skill references Google Drive document IDs and an interactive 'nlm login' flow but declares no credentials or auth requirements in the metadata — a mismatch between what it will actually need and what it declares.
Instruction Scope
The SKILL.md instructs the agent to run local GUI file pickers and arbitrary shell commands, clone a GitHub repo into ~/Programs and execute its run.sh (which initializes a virtualenv). That means downloading and executing third-party code on the user's machine. The instructions also direct polling with long sleeps and handling of user Google-authenticated resources (Drive) — all expected for the task but high-impact actions. There is no guidance to verify the safety or provenance of the cloned repo or run.sh.
Install Mechanism
No formal install spec in the skill metadata, but SKILL.md recommends running 'pip install notebooklm-mcp-cli' (reasonable) and cloning https://github.com/SPACESODA/epub2txt.git then running its run.sh. Downloading + executing code from a personal GitHub repo (unknown maintainer) is higher-risk than using an audited package or a well-known release host. The 'uv tool install' line is ambiguous and may not be applicable in all environments.
Credentials
The skill declares no required environment variables or primary credential, yet it requires interactive authentication ('nlm login') and expects access to Google Drive documents. Those are effectively privileged credentials and access to user data but are not represented in the metadata. The skill requests local filesystem access via GUI pickers and will read/upload local files — appropriate for the function but sensitive and should be explicit.
Persistence & Privilege
always:false and normal autonomous invocation are fine. The skill will create ~/Programs/epub2txt and initialize a virtualenv there (persistent files on disk), which is reasonable for epub conversion but is another write/execute action on the user's system. It does not request to change other skills' configs or system-wide settings.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install notebooklm-audio-generator - 安装完成后，直接呼叫该 Skill 的名称或使用
/notebooklm-audio-generator触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
notebooklm-audio-generator 1.0.0
- Automates uploading multiple files, URLs, YouTube links, Google Drive docs, or text to a NotebookLM notebook and generates a deep-dive audio overview in a preferred language.
- Guides the user step-by-step through notebook creation, source selection and addition, audio generation, and downloading via an intuitive flow.
- Supports .pdf, .txt, and .epub files (auto-converts .epub), and assists with GUI file/directory selection on all major platforms.
- Allows choosing the output audio language (supports en, zh, ja, es, fr, de) and destination folder using friendly pickers.
- Monitors generation progress and ensures the user receives the latest audio artifact.
- Verifies completion and downloaded file integrity, then notifies the user when done.
元数据
常见问题
NotebookLM Audio Generator 是什么？
Automates uploading multiple sources (files, URLs, YouTube, Drive, text) to a NotebookLM notebook, generating a deep dive audio overview in a preferred langu... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 367 次。
如何安装 NotebookLM Audio Generator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install notebooklm-audio-generator」即可一键安装，无需额外配置。
NotebookLM Audio Generator 是免费的吗？
是的，NotebookLM Audio Generator 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
NotebookLM Audio Generator 支持哪些平台？
NotebookLM Audio Generator 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 NotebookLM Audio Generator？
由 shiyanch（@shiyanch）开发并维护，当前版本 v1.0.0。
推荐 Skills