NotebookLM: Grounded Research & Studio Content

    1

    Ask questions grounded in your own sources, run web research, and generate podcasts, slides and quizzes in Google NotebookLM from your agent.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    NotebookLM: Grounded Research & Studio Content

    Example session with this skill installed

    Make a NotebookLM notebook from the Wikipedia articles on Photosynthesis and C4 carbon fixation. Then explain how C4 plants avoid photorespiration and why that matters in hot, dry climates, citing the sources. After that, give me three crops that use C4 photosynthesis and one that doesn't.

    • Read your context and instructions
    • Compiled the notebooklm

    I created a NotebookLM notebook with your two sources and asked it your questions. Every answer below comes only from those articles; the numbers in brackets are NotebookLM's citations to the passages it used.

    Notebook: 2 sources: Photosynthesis and C4 carbon fixation (Wikipedia)


    How C4 plants avoid photorespiration

    C4 plants avoid photorespiration by spatially concentrating carbon dioxide (CO₂) around the enzyme RuBisCO, eliminating energy waste and significantly boosting water efficiency in hot, arid environments [1-3].


    How C4 Plants Avoid Photorespiration

    The Photorespiration Problem: In standard C3 carbon fixation, the enzyme

    RuBisCO catalyses two competing reactions depending on the relative concentrations of CO₂ and O₂: carboxylation (incorporating CO₂) or oxygenation (binding O₂) [1, 4]. Oxygenation generates phosphoglycolate, a toxic compound that requires considerable plant energy to recycle through

    photorespiration [1, 4].
    2.

    Spatial Separation (Kranz Anatomy): C4 plants isolate RuBisCO from oxygen by using two partially isolated compartments in their leaves—outer

    mesophyll cells and inner

    bundle-sheath cells (an arrangement often called Kranz anatomy) [2, 5, 6].
    3.

    Primary Fixation by PEP Carboxylase: Inside mesophyll cells, CO₂ is first fixed by the enzyme

    PEP carboxylase (PEPC) onto phosphoenolpyruvate (PEP) to create a four-carbon organic acid (such as oxaloacetate, malate, or aspartate) [2, 7-9]. PEPC has a high affinity for carbon and is not confounded by O₂ [9].
    4.

    Concentrating CO₂ Around RuBisCO: These 4-carbon organic acids diffuse through plasmodesmata into the bundle-sheath cells, where they are decarboxylated to release CO₂ directly around RuBisCO [2, 6, 7, 9]. This localized, CO₂-rich environment saturates RuBisCO and suppresses photorespiration [1, 2, 6].


    Why This Matters in Hot, Dry Climates

    High Temperature Resilience: Warmer temperatures decrease gas solubility and dramatically increase photorespiration rates in C3 plants [3, 10]. The CO₂-concentrating mechanism overcomes reduced CO₂ solubility, sustaining high photosynthetic rates at elevated temperatures [3].
    *

    Lower Stomatal Conductance & Water Conservation: In hot and arid conditions, plants close their stomatal pores to reduce water evaporation [3, 11]. Because PEPC operates efficiently even at low stomatal conductance, C4 plants retain internal carbon gradients without losing excessive water [3].
    *

    Dramatic Water-Use Efficiency: When grown at 30 °C, C3 grasses lose approximately 833 water molecules for every CO₂ molecule fixed, whereas C4 grasses lose only about 277 [11]. This high water-use efficiency conserves soil moisture, allowing C4 plants to grow longer and colonize dry environments [11, 12].


    🌾 Would you like to explore the genetic engineering efforts currently underway to convert C3 crops (such as rice) to C4 photosynthesis?


    C4 crops (follow-up in the same conversation)

    Here are three major crops that use C4 photosynthesis and one key crop that does not:

    C4 Crops

    • Maize (corn) [1]
    • Sugarcane [1]
    • Sorghum [1] (Other C4 food crops include millet and teff [1])

    Non-C4 Crop (C3 Plant)

    Rice — Rice is naturally a

    C3 plant, which makes it less water-efficient in warmer environments compared to C4 crops [2].


    🌾 Would you like to delve deeper into the global scientific initiative aimed at genetically converting rice into a C4 crop?


    Next steps: I can generate an audio overview or a 5-question quiz from this notebook, or run a web research pass on C3-to-C4 engineering in rice and add the best results as sources. Studio content uses your NotebookLM quota, so I'll confirm before starting.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Turn Google NotebookLM into a tool your agent can use. Point it at your documents, papers or web pages and it answers only from those sources, with citations, instead of from its general training.

    What your agent can do

    • Build notebooks: create them and add web pages, YouTube videos, pasted text, local files or Google Drive documents as sources.
    • Ask your sources: get grounded answers with citations, follow up in the same conversation, or limit a question to specific documents.
    • Research for you: run a fast or deep web search on a topic, review what it found, and import the results you pick as new sources.
    • Generate Studio content: audio overviews (podcast-style), video overviews, slide decks (PDF), reports and study guides, infographics, flashcards and quizzes, in any language.

    How it works

    This skill tells your agent when to use NotebookLM and how to run each workflow safely. It confirms with you before anything that uses your NotebookLM quota or deletes data, and it knows how to recover when your session expires.

    It drives the open-source MCP server @m4ykeldev/notebooklm-mcp (MIT), which runs locally on your machine with your own Google session. No third-party service sits in between, and your cookies never leave your computer.

    Setup (about 2 minutes)

    1. npx -y @m4ykeldev/notebooklm-mcp auth: Chrome opens and waits while you sign in to Google, once.
    2. Register the server with your agent (one command for Claude Code; a JSON snippet for Cursor, VS Code and Claude Desktop is included).
    3. Copy the skill folder into your agent's skills directory.

    Requires Node.js 20+ and Google Chrome or Chromium.

    Requires Node.js 20+ and a one-time browser sign-in to Google (npx command). Deep research and Studio generation can take several minutes. Only web research results can be imported, not Drive ones. Long queries on large notebooks may need a higher --query-timeout. NotebookLM has no public API: the server uses the web app's internal endpoint, which Google can change without notice.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 27 days ago

    • Free with an Agensi account

    Requires

    Google Account
    NotebookLM MCP
    Node.js 20+
    Google Chrome or Chromium
    @m4ykeldev/notebooklm-mcp
    Node.js

    Permissions

    Terminal / Shell
    Read Files
    Network Access

    Allowed Hosts

    accounts.google.com

    File Scopes

    ~/.notebooklm-mcp/**

    Authentication cookies are stored only on your machine, in ~/.notebooklm-mcp/auth.json (mode 0600). The skill never asks for passwords. Network access is only to NotebookLM; local files are read only when you add one as a source.

    Listed28 days ago

    What's inside

    Frequently Asked Questions

    This skill solves the issue of fragmented context by allowing your AI agent to programmatically ingest local files, web links, and Google Drive docs into a centralized, grounded research notebook.

    The skill utilizes the Model Context Protocol (MCP) and is compatible with any AI agent or IDE that supports MCP, such as Claude Desktop or other modern agentic frameworks.

    Yes, you will need to perform a one-time manual authentication in your browser to secure session cookies, which allows the skill to securely access your notebooks and Google Drive data.

    This skill can generate a variety of specialized artifacts including podcast-style audio overviews, video explainers, study guides, slide decks, and quizzes directly from your sources.

    Yes. It can run a fast or deep web search on a topic, show you what it found, and import the results you choose as sources. Only web results can be imported; Drive search results are listed but not imported.

    New in Research & Analysis