REA
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Google Gemini Notebook (formerly NotebookLM) over MCP + a local HTTP REST API. Citation-backed Q&A, full Studio generation, chat history, multi-account rotation. For Claude Code, Codex, Cursor, n8n, Zapier, Make.
如果你是作者,可把这段代码贴到 README。徽章会链接回本页。
[](https://mcp.popnx.com/en/skills/roomi-fields-notebooklm-mcp)在同一工作中把 AI 连到服务的 MCP
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Codebase intelligence for TypeScript and JavaScript, exposed as structured MCP analysis tools.
MCP server for terminal commands, file operations, and process management
同类工作,排序与 Skill 列表相同
| 本项 notebooklm-mcp | rea | andrej-karpathy-skills | graphify | MoneyPrinterTurbo | |
|---|---|---|---|---|---|
| 作者 | roomi-fields | morluto | multica-ai | Graphify-Labs | harry0703 |
| Skill 数 | 1 | 1 | 1 | 1 | 1 |
| GitHub 星标 | 241 | 465,460 | 233,910 | 149,364 | 144,149 |
| 最近代码更新 | 2026年11月17日 | 2026年11月17日 | 2026年4月20日 | 2026年11月17日 | 2026年11月17日 |
| 许可证 | MIT | MIT | – | Apache-2.0 | MIT |
roomi-fields 提供的 1 个 Skill(指令集),AI 读取后照做 · 不是 MCP server
MCP = 手
把 AI 连到外部服务或数据,例如读取 Notion 或开发票。需要有程序或服务器在运行。
Skill = 手册
教 AI 怎么做某项工作的文本文件,不连接任何服务。
有些会重叠:少数 Skill 带有调用工具的脚本,少数 MCP 只提供知识。请从你想完成的工作出发。
Claude CodeClaude (เว็บ/แอป)CodexCursor
依据开发者 README 和 repo 中的文件夹
GitHub account roomi-fields (person) is only 11 months old: take extra care
★ 241 stars, few users so far · code last changed 2026年11月17日
1 scripts that can run on a computer: scripts/nblm.sh
Scripts that go online: scripts/nblm.sh
No hooks or writes to machine settings found
Checked automatically from the repo files at commit 65f7cd0 (read only, nothing was run). Not a safety guarantee: read it and decide for yourself
第一步:你用哪个 AI?
选择你用的应用,本页只显示该应用的安装方法
Uploading to claude.ai would fail: the description is over 200 characters (541). Use "Let the AI install it" instead
复制这句话,粘贴到 Claude Code 的对话中
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
重新打开 Claude Code,输入 /notebooklm,或直接提出相关任务,Claude 会自动选用
Checked against official docs 2026年11月12日 · Source: code.claude.com
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
正在检查命令 · 通过后才会显示复制按钮
摘自开发者 README,会下载并运行开发者的代码
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
ChatGPT 桌面应用内置 Codex,Codex 从 ~/.agents/skills 文件夹读取 Skill
在 ChatGPT 桌面应用中打开 Codex,把这句话粘贴到对话中
The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
把 notebooklm 文件夹放到 ~/.agents/skills/notebooklm
~ 是你的主文件夹($HOME)
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
Checked against official docs 2026年11月15日 · Source: learn.chatgpt.com
不能使用你自己安装的 Skill,只能使用目录中插件自带的 Skill
想在 ChatGPT 中使用这个 Skill,请按上面的方法使用 ChatGPT 桌面应用
Checked against official docs 2026年11月15日 · Source: learn.chatgpt.com
复制这句话,粘贴到 Codex 的对话中 · 适用于 ChatGPT 桌面应用中的 Codex、Codex CLI 和 IDE 扩展
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Codex 会自动发现新安装的 Skill;如果没有出现,请重启 Codex · 使用方法:在 ChatGPT 中输入 @ 并选择 Skill · 在 Codex CLI 或 IDE 扩展中运行 /skills,或输入 $ 加 Skill 名称
Checked against official docs 2026年11月15日 · Source: learn.chatgpt.com
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
复制这句话,粘贴到 Cursor 的对话中
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
在 Cursor 的 Agent 聊天中输入 / 并选择 notebooklm
Checked against official docs 2026年11月12日 · Source: cursor.com
Skill 是 AI 会照做的指令。安装前先阅读文件,只安装可信开发者的 Skill。
Mac:按 Command+空格输入 Terminal · Windows:下方命令在 PowerShell 中无法使用,请改用上方的“让 AI 帮你安装”
这条命令把 repo 下载到临时文件夹,只复制含 SKILL.md 的文件夹
正在检查命令 · 通过后才会显示复制按钮
适用于其他能读取 Skill 的应用。这句话列出了 Claude Code、Codex 和 Cursor 的路径,AI 会按你的应用选择 · Copy this sentence into the app’s chat
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Checked against official docs 2026年11月12日 · Source: code.claude.com, learn.chatgpt.com, cursor.com
其他支持 Agent Skills 的应用(如 Gemini CLI、GitHub Copilot)使用不同文件夹,请查看该应用的指南,再把含 SKILL.md 的文件夹复制过去。
已于 2026年11月11日 对照 Claude Code、Codex 和 Cursor 官方指南核实 skill 文件夹
以下文字来自开发者的 repo,不是 PopNx MCP 的指令
4,962 个字符 · 在 GitHub 打开完整文件
---
name: notebooklm
description: This skill should be used when the user wants to query their Google NotebookLM notebooks for citation-backed, source-grounded answers, or manage notebooks, sources, and Studio content (audio, report, video, infographic, presentation, data table, flashcards, quiz, mind map). It drives the @roomi-fields/notebooklm-mcp engine — via the notebooklm MCP tools when they are available in the session, otherwise via its HTTP REST API — and covers Google login, citation formats, the daily-quota-aware batch/ingestion pattern, and source discovery.
---
# NotebookLM
## Overview
NotebookLM answers questions **only from the sources uploaded to a notebook**,
with inline citations to the exact passages used — no open-web knowledge, so
answers are hallucination-resistant and fully traceable. This skill drives the
`@roomi-fields/notebooklm-mcp` engine to query notebooks, manage sources, and
generate Studio content, and encodes the patterns that make NotebookLM usable at
research scale (citation formats, the ~50-queries/day quota, batch-to-cache).
## Choosing the transport
Two ways reach the same engine — pick per what the session already has:
1. **notebooklm MCP tools** — if tools such as `notebook_ask` / `source_add` /
`server_health` (or `mcp__notebooklm__*`) are available in the session, call
them directly. This is the preferred path and needs no server.
2. **HTTP REST API** — otherwise, use the bundled `scripts/nblm.sh`, which talks
to a running NotebookLM MCP server (default `http://localhost:3000`,
override with `NOTEBOOKLM_SERVER_URL`). If no server is reachable, ask the
user to start one (`npm run start:http` from a clone) or to install the MCP.
Both are backed by the same account and session, so the choice is purely about
which is already wired up.
## Prerequisite: one Google login
NotebookLM needs a signed-in Google session (saved once, reused across runs).
Verify with `nblm.sh health` (or the `server_health` tool) — look for
`authenticated: true`. If not authenticated, run the interactive login **in a
terminal** (a visible Chrome window opens):
```bash
notebooklm-mcp-setup-auth # global install
# or: scripts/nblm.sh auth
```
Run the login in a terminal rather than through an in-client tool: interactive
Google login can take minutes and a stdio client's tool-call timeout may cut it
off.
## Core tasks
Use `scripts/nblm.sh` for the REST path (or the equivalent MCP tool):
```bash
scripts/nblm.sh health # reachability + auth status
scripts/nblm.sh notebooks # list notebooks (id + name)
scripts/nblm.sh ask "<question>" <notebook_id> # citation-backed answer (JSON citations)
scripts/nblm.sh search "<query>" <notebook_id> [limit] # ranked passages, no answer, no quota
scripts/nblm.sh generate <notebook_id> report # audio|report|video|infographic|presentation|data_table|flashcards|quiz|mind_map
```
- **Search before you ask.** `search` returns the indexed passages matching a
query, ranked best first, and **spends no chat quota** because nothing is
generated. When you intend to reason over the evidence yourself, it is both
cheaper and faster than asking. Rank is global across sources and lower is
better; `start`/`end` are offsets into that source's full text, and are null
when the server gave none.
- **Ask**: the script requests `source_format: json`, so the answer carries
source names + cited excerpts. For a human-facing answer, prefer `expanded`
(see `references/rest-api.md` to vary the format).
- **Generate**: `flashcards`/`quiz` route to the study-aid endpoint and
`mind_map` to the mind-map endpoint automatically.
## Working effectively (read before large runs)
For anything beyond a few questions, load `references/research-workflows.md`. Key
points:
- **Quota**: free accounts cap at ~50 chat queries/day. Rotate accounts
(`/re-auth`) or, better, **ingest once and retrieve offline**.
- **Batch → cache**: for literature reviews / SOTA surveys, run an exhaustive
question set through `/batch-to-vault` (writes markdown + `nblm-answer-v1` JSON
sidecars with citations), then answer repeated questions from the cache
(e.g. with [RTFM](https://github.com/roomi-fields/rtfm)) — unlimited, offline.
- **Fresh vs. follow-up**: omit `session_id` for independent questions (fastest);
pass a stable one to continue a conversation.
## References
- `references/rest-api.md` — endpoint + body reference for the HTTP path.
- `references/research-workflows.md` — citation formats, quota strategy, the
batch/ingestion pattern, source discovery.
## Installing the engine
If neither the MCP tools nor a server are present, the engine is the npm package
[`@roomi-fields/notebooklm-mcp`](https://github.com/roomi-fields/notebooklm-mcp)
(also a Claude Code plugin via the `roomi-fields/claude-plugins` marketplace).
Point the user there, then run the one-time login above.
提问、分享技巧或报告无法使用。按“有帮助”数量排序。