NotebookLM MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| NOTEBOOKLM_COOKIES | No | Cookie string to authenticate with NotebookLM. Set this to skip the auth flow in CI/headless environments. | |
| NOTEBOOKLM_CSRF_TOKEN | No | CSRF token for authentication. Optional if NOTEBOOKLM_COOKIES is set. | |
| NOTEBOOKLM_SESSION_ID | No | Session ID for authentication. Optional if NOTEBOOKLM_COOKIES is set. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| notebook_listA | List all NotebookLM notebooks with metadata (title, sources count, ownership) |
| notebook_createA | Create a new NotebookLM notebook |
| notebook_getB | Get details of a specific notebook including its sources |
| notebook_describeB | Get an AI-generated summary of the notebook content |
| notebook_renameB | Rename a notebook |
| notebook_deleteA | Delete a notebook (requires confirm=true) |
| source_describeA | Get metadata for a specific source in a notebook |
| source_get_contentA | Get the underlying text content of a source (used by grounding tool) |
| notebook_add_urlB | Add a website URL source to a notebook |
| notebook_add_textC | Add a text document source to a notebook |
| notebook_add_driveA | Add a Google Drive file source to a notebook |
| source_sync_driveC | Sync all Drive sources in a notebook to pull latest changes |
| source_deleteA | Delete a source from a notebook (requires confirm=true) |
| audio_overview_createC | Generate an audio podcast overview (requires confirm=true) |
| video_overview_createC | Generate a video overview (requires confirm=true) |
| infographic_createB | Generate an infographic (requires confirm=true) |
| slide_deck_createB | Generate a slide deck (requires confirm=true) |
| report_createC | Generate a text report (requires confirm=true) |
| flashcards_createC | Generate flashcards (requires confirm=true) |
| quiz_createC | Generate a quiz (requires confirm=true) |
| studio_statusC | Check the status of generated Studio artifacts |
| refresh_authA | Reload authentication tokens (re-extract CSRF and session from page) |
| save_auth_tokensB | Manually save authentication cookies (fallback method — prefer using CLI auth) |
| notebook_queryB | Ask a question about the sources in a notebook |
| research_startB | Start a web or Drive research task |
| research_statusC | Check the status of research tasks |
| research_importA | Import discovered sources from a research task into the notebook |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 27 tools
Each tool targets a distinct resource and action: notebook management, source operations, research tasks, artifact creation, query, and auth. Despite many artifact creation tools, each clearly specifies its output type (audio, video, infographic, etc.), so no two tools appear to do the same thing.
Most tools follow a resource_action pattern (e.g., notebook_create, source_delete, report_create), which is consistent. A few outliers like save_auth_tokens and refresh_auth use action_resource order, and research_status and studio_status are noun_noun compounds, but these are minor deviations within a largely coherent scheme.
At 27 tools, the server exceeds the 25-tool threshold for 'too many'. The count is inflated by 8 separate artifact creation tools that could be unified into a single tool with a type parameter, and the three add-source tools similarly could be consolidated. While the server covers a broad domain, the tool surface is heavier than necessary.
The tool set covers the full lifecycle for notebooks (create, read, list, rename, delete), sources (add three types, delete, describe, get content, sync), research (start, status, import), and artifacts (create and check status). Minor gaps exist such as no explicit list_sources tool (though notebook_get provides sources) and no update_source, but these are manageable workarounds.