{
"markdown": "\n\n# NotebookLM REST API + MCP server\n\n**Automate Google NotebookLM at scale. 33-endpoint HTTP REST API for n8n / Zapier / Make / curl, plus an MCP server for Claude Code / Cursor / Codex. Citation-backed Q&A, full Studio generation (audio · video · infographic · report · presentation · data table), multi-account rotation with auto-reauth across personal and Google Workspace accounts.**\n\n> v3.2.0 — **generated content can finally be deleted** (`content_delete`): the endpoint had been declared and called by nothing since v3, so notebooks accumulated every draft ever asked for. Builds on 3.1.x, where **generated content stopped coming back in the wrong language** — the interface locale was overriding the `language` argument on both transports, silently, while reporting success. Also: reading a source's full indexed text (`source_read`, paginated), working source labels, and RPC refusals reported as refusals instead of as a rotated endpoint id. Built on a **dual transport** — the internal `batchexecute` RPC API (10-100× faster than scraping, immune to UI rebrands) with the Playwright browser as an automatic fallback, both shipped permanently. Batch-tested on overnight runs of 1 000+ questions. See the [changelog](./CHANGELOG.md). [Compare with `PleasePrompto/notebooklm-mcp`](https://roomi-fields.github.io/notebooklm-mcp/compare) for when this project is the right pick (REST API, full Studio, auto-reauth).\n\n> **Note (July 2026):** Google rebranded **NotebookLM** to **Gemini Notebook**. It is the same product, existing links redirect, and this project drives the same underlying service — the browser path was updated for the new DOM in v2.3.0 and the RPC path in v3.0.0. Package and repository keep the `notebooklm` name.\n\n\n\n[](https://github.com/roomi-fields/notebooklm-mcp/actions/workflows/ci.yml) [](https://www.npmjs.com/package/@roomi-fields/notebooklm-mcp) [](https://www.npmjs.com/package/@roomi-fields/notebooklm-mcp) [](https://codecov.io/gh/roomi-fields/notebooklm-mcp) [](https://opensource.org/licenses/MIT) [](https://www.typescriptlang.org/) [](https://nodejs.org/)\n\n[](https://modelcontextprotocol.io/) [](https://claude.ai/claude-code) [](./deployment/docs/04-N8N-INTEGRATION.md) [](https://github.com/roomi-fields/notebooklm-mcp) [](https://mcptoplist.com/server/io.github.roomi-fields%2Fnotebooklm-mcp)\n\n\n\n
\n\n---\n\n> **Unofficial project — good to know before you start**\n>\n> This is not affiliated with Google. It talks to the same `batchexecute`\n> endpoints the NotebookLM web app uses, with a browser fallback when they move.\n> They are undocumented, so they can change without notice — when that happens we\n> ship a fix, as we have for every change so far.\n>\n> Two practical notes: use a **dedicated Google account** for automation, and\n> expect NotebookLM's own quotas to apply at high volume. See\n> [Disclaimer](#disclaimer) for the full text.\n\n---\n\n## What You Can Build\n\n🔗 **No-code automation pipelines** — The 33-endpoint REST API means NotebookLM becomes a step in n8n, Zapier, Make, or a plain `curl` in cron. No agent, no MCP client, no Node in your stack — just HTTP. This is the half most NotebookLM libraries don't have.\n\n🤖 **Agent tooling** — The same engine over MCP for Claude Code, Cursor and Codex, with a bundled skill that primes the agent on citation formats, the daily-quota-aware batch pattern, and transport selection.\n\n📚 **Research at volume** — Multi-account rotation with automatic re-authentication, built for overnight runs of 1 000+ questions across several notebooks without babysitting.\n\n🎙️ **Full Studio generation** — Audio overviews, video, infographics, reports, presentations, data tables, plus flashcards, quizzes and mind maps — generated and downloaded programmatically.\n\n## Use Cases & Recipes\n\nNotebookLM is a **grounded** engine: Gemini reads your sources and answers _from them_, with citations. The winning pattern is to let it do the expensive reading while your own stack handles orchestration and the last mile.\n\n**Spend fewer tokens — offload the reading**\n\n- **🪙 Zero-token synthesis layer** — Drop 30 documents in a notebook, let Gemini do the heavy analysis, and spend your agent's context only on the final polish. The reasoning happens server-side; your agent just orchestrates (`add_notebook` → `source_add` → `notebook_ask`).\n- **💾 Answer cache you can re-read offline** — `vault_batch` writes every answer to disk as structured JSON against a published [schema](https://schemas.roomi-fields.com/nblm-answer-v1.json), so a batch run becomes a corpus you can grep, diff, re-index, or feed to a retrieval layer — without re-querying and re-spending quota.\n\n**Wire it into things that aren't agents**\n\n- **⚙️ NotebookLM as an n8n / Zapier / Make step** — Because it speaks plain HTTP, a citation-backed answer becomes one node in a workflow: a form submission triggers a question, the cited answer lands in a sheet, a Slack message, or a database. No agent runtime involved.\n- **📄 Document intake pipeline** — Watch a folder or an inbox, push new PDFs and URLs in as sources, and ask a standing set of questions against them on every arrival.\n\n**Grounded answers with a paper trail**\n\n- **🔍 Citations with the actual source text** — Answers come back with source names _and the quoted excerpts_ they rest on, extracted from the citation panel — so a claim can be checked, not just attributed.\n- **🎓 Literature review at thesis scale** — Batch 100+ research questions across multiple notebooks, rotate accounts as daily quotas run out, and resume where it stopped. Built for, and tested on, exactly this.\n\n**Get artifacts back out**\n\n- **🔁 One source set, every format** — Fan a single notebook out to a podcast, a video, a slide deck, a report, a quiz and a mind map, then download them all locally.\n\n---\n\n## In the Wild\n\nReal deployments, not hypotheticals.\n\n- **📚 A doctoral literature review at batch scale** — The project was built for, and is\n continuously tested on, overnight runs of **1 000+ research questions** spread across\n several notebooks: multi-account rotation picks up when a daily quota runs out, every\n answer is written to disk with its citations, and an interrupted run resumes instead of\n starting over. The batch pattern in [`vault_batch`](#features) exists because a thesis\n needed it.\n\n- **🔌 Replacing a RAG engine with the REST API** — [`musnymubarak/Calim_Doc`](https://github.com/musnymubarak/Calim_Doc)\n swapped a Gemini-based retrieval engine for this project's HTTP API, running it as a\n Docker service (`notebooklm:3000`) behind a full client and worker layer. A good\n illustration of the REST half: no agent runtime, no MCP client — NotebookLM simply\n became a backend service their Python app calls.\n\nBuilt something with it? Open an issue — this section is for other people's work.\n\n## Features\n\n### Q&A with Citations\n\n- **Ask questions** to NotebookLM and get accurate, citation-backed answers\n- **Source citation extraction** with 5 formats: none, inline, footnotes, json, expanded (97% excerpt success rate)\n- **Session management** for multi-turn conversations with auto-reauth on session expiry\n\n### Content Generation\n\nGenerate multiple content types from your notebook sources:\n\n| Content Type | Formats | Options |\n| ------------------ | ------------------------ | ---------------------------------------------- |\n| **Audio Overview** | Podcast-style discussion | Language (80+), custom instructions |\n| **Video** | Brief, Explainer | 6 visual styles, language, custom instructions |\n| **Infographic** | Horizontal, Vertical | Language, custom instructions |\n| **Report** | Summary, Detailed | Language, custom instructions |\n| **Presentation** | Overview, Detailed | Language, custom instructions |\n| **Data Table** | Simple, Detailed | Language, custom instructions |\n| **Flashcards** | Study cards | Language, custom instructions |\n| **Quiz** | Assessment questions | Language, custom instructions |\n| **Mind Map** | Interactive node graph | Saved to the notebook |\n\n**Video Visual Styles**: classroom, documentary, animated, corporate, cinematic, minimalist\n\n**Language of generated content**: pass `language` to any generator — a BCP-47 code (`es`, `ja`, `pt_BR`, `zh_Hans`) or a name in English or in the language itself (`\"Spanish\"`, `\"Español\"`). 81 languages are accepted, and an unrecognised one is refused rather than quietly swapped for another. Set a default with `NOTEBOOKLM_CONTENT_LANGUAGE`; it is deliberately independent of `NOTEBOOKLM_UI_LOCALE`, which only picks the interface language the browser fallback reads.\n\nFlashcards and quizzes are generated via `generate_study_aid`; mind maps via `generate_mind_map`. v3 also adds `share_notebook`, `manage_labels`, and `research_sources` (web/Drive source discovery) — see the [changelog](./CHANGELOG.md).\n\n### Content Download\n\n- **Download Audio** — WAV audio files\n- **Download Video** — MP4 video files\n- **Download Infographic** — PNG image files\n- Text-based content (report, presentation, data_table) is returned in the API response\n- **Delete generated content** (`content_delete`) — until now a notebook accumulated every draft anyone ever asked for, with no way to remove one short of the web UI\n\n### Source Management\n\n- **Add sources**: Files (PDF, TXT, DOCX), URLs, Text, YouTube videos, Google Drive\n- **List sources**: Every source with its ID and title (`source_list`)\n- **Read a source in full** (`source_read`): the exact text NotebookLM indexed — what it actually reasons over, which the web UI only shows in fragments. Quote a source verbatim, check what a PDF really yielded, or hand the raw material to another tool. Name the source instead of its ID if you prefer; an ambiguous name is refused rather than guessed. Long sources arrive **one page at a time**, with an explicit instruction for fetching the next — or `paginate: false` for the whole document at once.\n\n### Notebook Library\n\n- **Multi-notebook management** with validation and smart selection\n- **Auto-discovery**: Automatically generate metadata via NotebookLM queries\n- **Search notebooks** by keyword in name, description, or topics\n- **Scrape notebooks**: List all notebooks from NotebookLM with IDs and names\n- **Bulk delete**: Delete multiple notebooks at once\n\n### Accounts & Localization\n\n- **Personal _and_ Google Workspace accounts** — recognizes both NotebookLM hosts (`notebooklm.google.com` and the `notebook.google.com` Workspace alias), so Workspace sessions authenticate cleanly instead of looping on \"session expired\"\n- **UI-language-aware** — drives NotebookLM whether its interface is in English, French, German, or Japanese (`en` · `fr` · `de` · `ja`); [add a language](./docs/ADDING_A_LANGUAGE.md) in a single JSON file\n\n### Integration Options\n\n- **MCP Protocol** — Claude Code, Cursor, Codex, any MCP client\n- **Agent Skill** — ships a bundled [`notebooklm` skill](./skills/notebooklm/SKILL.md) (also standalone: [`roomi-fields/notebooklm-skill`](https://github.com/roomi-fields/notebooklm-skill)) that teaches the agent citation formats, the daily-quota-aware batch pattern, and when to use which transport\n- **HTTP REST API** — n8n, Zapier, Make.com, custom integrations\n- **Docker** — Isolated deployment with Docker or Docker Compose\n- **[RTFM](https://github.com/roomi-fields/rtfm) retrieval layer** — `/batch-to-vault` writes citation-backed answers as markdown + JSON sidecars (`nblm-answer-v1` schema), indexable by [RTFM](https://github.com/roomi-fields/rtfm) (FTS5 + semantic) for unlimited offline queries. Ideal for academic / SOTA workflows. [Guide](./deployment/docs/14-RTFM-INTEGRATION.md).\n\n---\n\n## Quick Start\n\n### Option 0 — Claude Code marketplace (one-liner, recommended for Claude Code users)\n\nThe fastest way to get NotebookLM into Claude Code. Distributed via the [`roomi-fields/claude-plugins`](https://github.com/roomi-fields/claude-plugins) marketplace alongside [RTFM](https://github.com/roomi-fields/rtfm) (the retrieval companion — see [RTFM integration guide](./deployment/docs/14-RTFM-INTEGRATION.md)):\n\n```text\n/plugin marketplace add roomi-fields/claude-plugins\n/plugin install notebooklm@roomi-fields\n```\n\nThat registers the MCP server, runs `npx -y @roomi-fields/notebooklm-mcp@` automatically (Node ≥ 18 required), and lets you upgrade with two commands when a new release ships: `/plugin marketplace update roomi-fields` then `/reload-plugins`. Then run `npx -y -p @roomi-fields/notebooklm-mcp notebooklm-mcp-setup-auth` once in a terminal to log into Google (a visible Chrome opens). To install RTFM at the same time: `/plugin install rtfm@roomi-fields`.\n\n### Option 1 — HTTP REST API (n8n, Zapier, Make, curl, any HTTP client)\n\n```bash\ngit clone https://github.com/roomi-fields/notebooklm-mcp.git\ncd notebooklm-mcp\nnpm install && npm run build\nnpm run setup-auth # One-time Google login\nnpm run start:http # Start REST API on port 3000\n```\n\n```bash\n# Citation-backed Q&A, single curl, JSON response\ncurl -X POST http://localhost:3000/ask \\\n -H 'Content-Type: application/json' \\\n -d '{\"question\": \"Summarize chapter 3\", \"notebook_id\": \"your-id\", \"source_format\": \"json\"}'\n```\n\nThe full surface is **33 documented endpoints** — see the [REST API reference](https://roomi-fields.github.io/notebooklm-mcp/notebooklm-rest-api). For overnight batches of 1 000+ questions, see the [batch pattern](https://roomi-fields.github.io/notebooklm-mcp/batch-1000-questions).\n\n### Option 2 — MCP Mode (Claude Code, Cursor, Codex)\n\n```bash\n# Build (same package, MCP transport)\ngit clone https://github.com/roomi-fields/notebooklm-mcp.git\ncd notebooklm-mcp\nnpm install && npm run build\n\n# Claude Code\nclaude mcp add notebooklm node /path/to/notebooklm-mcp/dist/index.js\n\n# Cursor — add to ~/.cursor/mcp.json\n{\n \"mcpServers\": {\n \"notebooklm\": {\n \"command\": \"node\",\n \"args\": [\"/path/to/notebooklm-mcp/dist/index.js\"]\n }\n }\n}\n```\n\n**Log in once — in a terminal, not through the assistant.** Run the interactive\nGoogle login as a command; a visible Chrome window opens, you sign in, and the\nsaved session is then reused by the MCP server:\n\n```bash\nnpm run setup-auth # from a clone (Option 2 above)\nnotebooklm-mcp setup-auth # from a global install (npm i -g @roomi-fields/notebooklm-mcp)\n```\n\nDo the login in a terminal rather than by asking the assistant _\"log me in\"_:\nsome stdio MCP clients (e.g. Claude Desktop) cap tool-call duration and cut off\nthe up-to-10-minute interactive login before you can finish signing in (see\n[issue #27](https://github.com/roomi-fields/notebooklm-mcp/issues/27)).\n\n### Option 3 — Docker (NAS, server, headless)\n\n```bash\n# Build and run\ndocker build -t notebooklm-mcp .\ndocker run -d --name notebooklm-mcp -p 3000:3000 -p 6080:6080 -v notebooklm-data:/data notebooklm-mcp\n\n# Authenticate via noVNC\n# 1. Open http://localhost:6080/vnc.html\n# 2. Run: curl -X POST http://localhost:3000/setup-auth -d '{\"show_browser\":true}'\n# 3. Login to Google in the VNC window\n```\n\nSee [Docker Guide](./deployment/docs/08-DOCKER.md) for NAS deployment (Synology, QNAP).\n\n---\n\n## Documentation\n\nFull docs site: **** · [OpenAPI 3.1 spec](./deployment/docs/openapi.yaml)\n\n| Guide | Description |\n| -------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------- |\n| [Installation](./deployment/docs/01-INSTALL.md) | Step-by-step setup for HTTP and MCP modes |\n| [Configuration](./deployment/docs/02-CONFIGURATION.md) | Environment variables and security |\n| [REST API reference](./deployment/docs/03-API.md) | Complete HTTP endpoint documentation (33 endpoints) |\n| [Run 1 000 questions overnight](./deployment/docs/12-BATCH-1000.md) | Production batch pattern with auto-reauth and rotation |\n| [**RTFM integration — cache as searchable vault**](./deployment/docs/14-RTFM-INTEGRATION.md) | Pipeline pattern: NotebookLM as one-shot ingestion, RTFM as retrieval layer. `/batch-to-vault` endpoint, `nblm-answer-v1` schema. |\n| [n8n integration](./deployment/docs/04-N8N-INTEGRATION.md) | Workflow automation setup |\n| [Troubleshooting](./deployment/docs/05-TROUBLESHOOTING.md) | Common issues and solutions |\n| [Notebook library](./deployment/docs/06-NOTEBOOK-LIBRARY.md) | Multi-notebook management |\n| [Auto-discovery](./deployment/docs/07-AUTO-DISCOVERY.md) | Autonomous metadata generation |\n| [Content management](./deployment/docs/10-CONTENT-MANAGEMENT.md) | Audio, video, infographic, report, presentation |\n| [Multi-account rotation](./deployment/docs/11-MULTI-ACCOUNT.md) | Multiple accounts with TOTP auto-reauth |\n| [Docker](./deployment/docs/08-DOCKER.md) | Docker and Docker Compose deployment |\n| [Multi-interface](./deployment/docs/09-MULTI-INTERFACE.md) | Run Claude Desktop + HTTP simultaneously |\n| [**Compare with PleasePrompto v2.0.0**](./deployment/docs/13-COMPARE.md) | Feature matrix vs the upstream MCP-only server |\n| [Chrome profile limitation](./docs/CHROME_PROFILE_LIMITATION.md) | Profile locking (solved in v1.3.6+) |\n| [Adding a language](./docs/ADDING_A_LANGUAGE.md) | i18n system for multilingual UI support |\n\n---\n\n## Roadmap\n\nSee [ROADMAP.md](./ROADMAP.md) for planned features and version history.\n\n**Latest releases:**\n\n- **v3.0.1** — Interactive Google login as a first-class CLI command (`notebooklm-mcp setup-auth`) for global / stdio-client installs; `setup_auth` / `re_auth` accept a top-level `headless` (#27)\n- **v3.0.0** — Major refactor: **dual transport** (NotebookLM's internal `batchexecute` RPC API with automatic DOM fallback), 10-100× faster and immune to UI rebrands; **5 new tools** (notebook sharing, study aids, mind maps, source labels, web research)\n- **v2.3.0** — Full support for Google's **\"Gemini Notebook\" rebrand**: create / list / rename / delete, sources, and every Studio generation type re-verified end-to-end (#23, #21)\n- **v2.2.1** — Recognize both NotebookLM hosts so **Google Workspace** accounts authenticate (the `notebook.google.com` alias); notebook listing no longer wastes ~30s after the \"Gemini Notebook\" rebrand; HTTP banner reads the real version. Diagnosis + patch by @kpietkaa (#19)\n- **v2.2.0** — Fix new-answer detection timing out when an answer repeats an earlier one (position-based identity, not text-hash); graceful shutdown on stdio disconnect; **Japanese** UI locale\n- **v2.1.1** — Thai UI selectors for `notebook_create` (partial, #18)\n- **v2.1.0** — `note_list` and `note_get` MCP tools (#17)\n- **v2.0.4** — German UI selectors (closes #14)\n- **v2.0.0** — Tools renamed to a namespaced tree (`notebook_ask`, `source_add`, `session_list`, `server_health`, `vault_batch`…) across 9 namespaces; `tools/list` advertises only the canonical names. **Backward compatible — the legacy flat names still work as aliases**, so existing scripts and configs keep running. Also adds MCP `annotations` (read-only / destructive / idempotent / open-world hints) and `outputSchema` + `structuredContent` on every tool. Published on the [Smithery registry](https://smithery.ai/servers/roomifields/notebooklm-mcp).\n- **v1.7.0** — `batch_to_vault` exposed as a first-class MCP tool (parity with the HTTP endpoint, no localhost server required); shared `runBatchToVault` helper deduplicates the loop across both transports\n- **v1.6.0** — `/batch-to-vault` endpoint + RTFM integration (`nblm-answer-v1` JSON Schema published at [schemas.roomi-fields.com/nblm-answer-v1.json](https://schemas.roomi-fields.com/nblm-answer-v1.json)) for caching NotebookLM answers as a searchable markdown vault\n- **v1.5.8** — NotebookLM 2026 UI adaptations (icon-label sanitization, Discussion-panel recovery, count-based source detection) — PR #5 by @KhizarJamshaidIqbal\n- **v1.5.7** — Citation extraction selector fix (`.highlighted`) and Docker multi-stage build — PR #1 by @JulienCANTONI\n- **v1.5.6** — Citation extraction major rewrite (97% success rate), browser-verified auth at startup, profile auto-sync\n- **v1.5.0** — Complete Studio content generation (video, infographic, presentation, data_table) + Notes management + Delete sources\n- **v1.4.0** — Content management (sources, audio, generation) + Multi-account\n\n_Intermediate patch and hardening releases (1.5.x–1.7.x) are in the full [CHANGELOG](./CHANGELOG.md)._\n\n**Not yet implemented:**\n\n- Discover sources (Web/Drive search with Fast/Deep modes)\n- Edit notes (create, delete, and convert are implemented)\n\n---\n\n## Disclaimer\n\nThis tool automates browser interactions with NotebookLM. Use a dedicated Google account for automation. CLI tools like Claude Code can make mistakes — always review changes before deploying.\n\nSee full [Disclaimer](#disclaimer-details) below.\n\n---\n\n## Contributing\n\nFound a bug? Have an idea? [Open an issue](https://github.com/roomi-fields/notebooklm-mcp/issues) or submit a PR!\n\nSee [CONTRIBUTING.md](./CONTRIBUTING.md) for guidelines.\n\n## License\n\nMIT — Use freely in your projects. See [LICENSE](./LICENSE).\n\n## Author\n\n**Romain Peyrichou** — [@roomi-fields](https://github.com/roomi-fields)\n\n## Acknowledgments\n\nThanks to everyone who has contributed code, ideas, and bug reports:\n\n- **Khizar Jamshaid Iqbal** — [@KhizarJamshaidIqbal](https://github.com/KhizarJamshaidIqbal), 2025 UI selector fixes, doctor script, PII scrub\n- **Kazik Pietka** — [@kpietkaa](https://github.com/kpietkaa), `notebook.google.com` rebrand support\n- **Rui Ruiberriz** — [@Excauboi](https://github.com/Excauboi), `hl=` on app URLs + click-through scrape fallback\n- **[@he0xwhale](https://github.com/he0xwhale)** — `note_list` / `note_get` MCP tools\n- **[@eminsnow](https://github.com/eminsnow)** — canonical tool names (`_` over `.`)\n- **Julien Cantoni** — [@JulienCANTONI](https://github.com/JulienCANTONI)\n\n---\n\n\nFull Disclaimer
\n\n**About browser automation:**\nWhile I've built in humanization features (realistic typing speeds, natural delays, mouse movements), I can't guarantee Google won't detect or flag automated usage. Use a dedicated Google account for automation.\n\n**About CLI tools and AI agents:**\nCLI tools like Claude Code, Codex, and similar AI-powered assistants are powerful but can make mistakes:\n\n- Always review changes before committing or deploying\n- Test in safe environments first\n- Keep backups of important work\n- AI agents are assistants, not infallible oracles\n\nI built this tool for myself and share it hoping it helps others, but I can't take responsibility for any issues that might occur. Use at your own discretion.\n\n \n\n---\n\n\n\nBuilt with frustration about hallucinated APIs, powered by Google's NotebookLM\n\n⭐ [Star on GitHub](https://github.com/roomi-fields/notebooklm-mcp) if this saves you debugging time!\n\n
\n",
"bytes": 26246,
"sha": "9cec9b7722aa3ffd21ee4af1a81dc62ffb113eaf8533203a7b6f734ef4948b3a",
"repo_slug": "roomi-fields/notebooklm-mcp",
"fonte": "repo",
"truncated": false,
"api": "https://agentalog.com/api/listings/mcp_io_github_roomi_fields_notebooklm_mcp_3300c8c4/readme"
}