Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/jacob-bd/notebooklm-mcp-cli/llms.txt

Use this file to discover all available pages before exploring further.

The management commands cover everything outside of notebooks and content: persisted configuration settings, short-name aliases for long IDs, per-notebook chat configuration, note management, diagnostics, artifact export, and MCP server setup for AI tool integrations.

​
Configuration

The nlm config commands read and write settings stored in ~/.notebooklm-mcp-cli/config.toml. All settings use a section.key dot-notation.

​
config show

Print the entire current configuration. 
nlm config show
nlm config show --json
Flags
FlagShortDescription
--json-jOutput as JSON instead of TOML

​
config get

Read a single configuration value. 
nlm config get auth.default_profile
nlm config get output.format
nlm config get output.color

​
config set

Write a single configuration value. Changes take effect immediately for subsequent commands. 
nlm config set auth.default_profile work
nlm config set output.format json
nlm config set output.color false
nlm config set output.short_ids true
Available settings
KeyDefaultDescription
output.formattableDefault output format: table or json
output.colortrueEnable coloured terminal output
output.short_idstrueShow shortened IDs in table output
auth.browserautoPreferred browser for login: auto, chrome, arc, brave, edge, chromium, vivaldi, opera. Falls back to auto if the chosen browser is not found.
auth.default_profiledefaultProfile used when --profile is not specified. The MCP server always uses the active default profile — changing this setting instantly switches the MCP server’s Google account.

​
Aliases

Aliases map short, memorable names to full notebook, source, or artifact UUIDs. Once set, any command that accepts an ID will also accept the alias.

​
alias set

Create or update an alias. The type (notebook, source, artifact, task) is auto-detected by default. 
nlm alias set myproject <notebook-id>
nlm alias set main-audio <artifact-id>
Flags
FlagShortDescription
--type-tForce type: notebook, source, artifact, task
--profile-pProfile to use for ID type detection

​
alias list

Show all defined aliases with their resolved IDs and types. 
nlm alias list

​
alias get

Resolve an alias to its full UUID. 
nlm alias get myproject
# → abc12345-6789-...

​
alias delete

Remove an alias. This does not affect the underlying notebook or artifact. 
nlm alias delete myproject
nlm alias delete myproject --confirm   # skip prompt

​
Using aliases

Once an alias is defined it works anywhere an ID is accepted: 
# All equivalent after: nlm alias set research <notebook-id>
nlm notebook get research
nlm source list research
nlm audio create research --confirm
nlm notebook query research "What are the key findings?"

​
Chat configuration

Customise the AI chat behaviour for a specific notebook — choose the goal template and response length, or provide a fully custom system prompt.

​
chat configure

# Use defaults
nlm chat configure <notebook-id> --goal default --response-length default

# Learning-focused, longer responses
nlm chat configure <notebook-id> --goal learning_guide --response-length longer

# Fully custom system prompt
nlm chat configure <notebook-id> \
  --goal custom \
  --prompt "You are an expert patent attorney. Answer all questions with reference to relevant case law."
Flags
FlagShortDescription
--goal-gChat goal template: default, learning_guide, custom
--response-length-rResponse length preference: default, longer, shorter
--promptCustom system prompt (required when --goal custom, max 10,000 chars)
--profile-pProfile to use
Goal templates
GoalDescription
defaultBalanced, general-purpose assistant
learning_guideSocratic style optimised for study and comprehension
customFree-form system prompt defined by --prompt

​
Notes

Notebooks support attached notes — short text snippets that are stored alongside sources and included in queries. The following commands manage notes inside a notebook. 
nlm note list <notebook-id>
nlm note list <notebook-id> --json
nlm note list <notebook-id> --quiet   # IDs only
Note command flags
CommandFlagShortDescription
note list--json-jOutput as JSON
note list--quiet-qOutput IDs only
note create--content-cNote content (required)
note create--title-tNote title (default: New Note)
note update--content-cNew content
note update--title-tNew title
note delete--confirm-ySkip confirmation prompt
nlm note delete is irreversible. The note is permanently removed. Use --confirm to skip the interactive prompt in scripts.

​
Export

Export studio artifacts to Google Docs or Google Sheets for further editing or sharing.

​
export artifact

Export any studio artifact to Google Docs (reports, study guides, etc.) or Google Sheets (data tables). 
# Export a report to Google Docs
nlm export artifact <notebook-id> <artifact-id> --type docs

# Export a data table to Google Sheets with a custom title
nlm export artifact <notebook-id> <artifact-id> --type sheets --title "My Data Table"
Flags
FlagShortDescription
--type-tExport destination: docs or sheets (required)
--titleTitle for the exported document
--json-jOutput result as JSON
--profile-pProfile to use

​
export to-docs

Shorthand for exporting a report artifact to Google Docs. 
nlm export to-docs <notebook-id> <artifact-id>
nlm export to-docs <notebook-id> <artifact-id> --title "Q3 Briefing"

​
export to-sheets

Shorthand for exporting a data table artifact to Google Sheets. 
nlm export to-sheets <notebook-id> <artifact-id>
nlm export to-sheets <notebook-id> <artifact-id> --title "Sales Data"

​
Doctor diagnostics

nlm doctor runs a suite of checks to verify that your installation, authentication, browser, and AI tool integrations are all working correctly. Run it whenever you encounter unexpected behaviour or after a fresh install. 
# Run all checks
nlm doctor

# Include additional detail (Python version, file paths, profile info)
nlm doctor --verbose
Checks performed
CategoryWhat is checked
InstallationPackage version, nlm and notebooklm-mcp binary paths
AuthenticationProfile existence, cookies present, CSRF token, associated Google account email
BrowserChromium-based browser found on PATH, saved Chrome profile for headless re-auth
AI ToolsMCP configuration status for each supported client (Claude Code, Gemini, Cursor, Windsurf)
Each failed check includes a suggested fix, for example: 
  Authentication: Cookies: missing
  → Run nlm login to authenticate
Flags
FlagShortDescription
--verbose-vShow Python version, file paths, last-validated time, and per-profile detail

​
Setup — MCP server configuration

nlm setup configures the NotebookLM MCP server transport for AI tools that support the Model Context Protocol. This is a one-time setup step per tool. 
# Add the MCP server to a specific tool
nlm setup add claude-code
nlm setup add gemini
nlm setup add cursor
nlm setup add windsurf
nlm setup add github-copilot

# Remove the server from a tool
nlm setup remove gemini

# List all supported clients and their current config status
nlm setup list

# Generate a JSON snippet for tools not in the registry
nlm setup add json
Supported clients: claude-code, gemini, github-copilot, cursor, windsurf, cline, antigravity, codex, opencode
nlm setup configures the MCP server transport layer. To install skill / reference documentation for AI coding tools that do not use MCP, use nlm skill install instead. See the MCP overview page for a full walkthrough of both.

Build docs developers (and LLMs) love

Get started for freeTalk to us