NotebookLM MCP

Ask the notebook

notebook_query

Ask NotebookLM a question restricted to a notebook's sources, then receive the answer and a conversation ID for follow-ups.

Instructions

Ask NotebookLM a question that it answers only from the notebook's sources (all of them, or source_ids). Returns the answer and a conversation_id. The answer text keeps NotebookLM's bracketed citation numbers such as [1], but not which passage each number points to. Pass conversation_id back to ask a follow-up; this server remembers the conversation's history while it keeps running.

Input Schema

Name	Required	Description
`query`	Yes	The question, in natural language
`source_ids`	No	Answer only from these source IDs (from notebook_get); omit to use every source
`notebook_id`	Yes	Notebook ID, as returned by notebook_list or notebook_create
`conversation_id`	No	conversation_id returned by a previous notebook_query, to ask a follow-up in the same conversation

Output Schema

Name	Required	Description
`answer`	Yes	NotebookLM's answer, with bracketed citation numbers such as [1]
`status`	Yes
`message`	No	Human-readable outcome or next step
`conversation_id`	Yes	Pass back as conversation_id to ask a follow-up

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv0.5.0Oct 20, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": false,
+  "properties": {
+    "answer": {
+      "description": "NotebookLM's answer, with bracketed citation numbers such as [1]",
+      "type": "string"
+    },
+    "conversation_id": {
+      "description": "Pass back as conversation_id to ask a follow-up",
+      "type": [
+        "string",
+        "null"
+      ]
+    },
+    "message": {
+      "description": "Human-readable outcome or next step",
+      "type": "string"
+    },
+    "status": {
+      "const": "success",
+      "type": "string"
+    }
+  },
+  "required": [
+    "status",
+    "answer",
+    "conversation_id"
+  ],
+  "type": "object"
+}

Changed4 schema fields changedv0.4.0Oct 20, 2026

changedInput schema / properties / conversation_id / description

Previous value: -"Conversation ID for follow-up questions"New value: +"conversation_id returned by a previous notebook_query, to ask a follow-up in the same conversation"

changedInput schema / properties / notebook_id / description

Previous value: -"The notebook ID"New value: +"Notebook ID, as returned by notebook_list or notebook_create"

changedInput schema / properties / query / description

Previous value: -"Question to ask"New value: +"The question, in natural language"

changedInput schema / properties / source_ids / description

Previous value: -"Specific source IDs to query (omit for all)"New value: +"Answer only from these source IDs (from notebook_get); omit to use every source"

First observedv0.3.1Aug 6, 2026

TDQS

A4.7/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds substantial context beyond the annotations: it discloses that the answer keeps bracketed citation numbers like [1] but loses the passage mapping, that it returns a conversation_id, and that server-side conversation memory is ephemeral ("while it keeps running"). The readOnlyHint=false is consistent with creating conversation state, so there is no contradiction.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three dense sentences, front-loaded with the core action and scope, then return shape, then the follow-up mechanism. No filler or repetition of the schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists so return fields need not be re-explained, and the description covers the remaining gaps: source scoping, citation limitation, and conversation persistence. An agent has everything needed to invoke it correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3. The description still adds meaning: it clarifies the source_ids scoping behavior (all sources otherwise) and explains the follow-up role of conversation_id, going slightly beyond the per-field schema text.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (ask a question) and resource (NotebookLM notebook) with an explicit scope restriction: it answers only from the notebook's sources, all of them or source_ids. This distinguishes it from siblings like notebook_summarize or research_start without opening their schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear operational context: restrict answers to all sources or specific source_ids, and pass conversation_id back to continue a conversation. However, it never explicitly contrasts this with alternative tools (e.g., notebook_summarize), so the agent must infer when Q&A is preferred over other retrieval options.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.