Programmatic access to Gemini Notebook - via command-line interface (CLI), Model Context Protocol (MCP) server, and AI agent skills.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add notebooklm -- uvx notebooklm-mcp-server~/.codex/config.toml
[mcp_servers.notebooklm]
command = "uvx"
args = ["notebooklm-mcp-server"]
# codex mcp add notebooklm -- uvx notebooklm-mcp-server~/.cursor/mcp.json
{
"mcpServers": {
"notebooklm": {
"command": "uvx",
"args": [
"notebooklm-mcp-server"
]
}
}
}.vscode/mcp.json
{
"servers": {
"notebooklm": {
"type": "stdio",
"command": "uvx",
"args": [
"notebooklm-mcp-server"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"notebooklm": {
"command": "uvx",
"args": [
"notebooklm-mcp-server"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"notebooklm": {
"command": "uvx",
"args": [
"notebooklm-mcp-server"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"notebooklm": {
"command": "uvx",
"args": [
"notebooklm-mcp-server"
]
}
}
}