nblm-rs

Unofficial NotebookLM Enterprise API client

🦀 Rust CLI: Command-line tool for shell scripting and automation
🐍 Python SDK: Python bindings for integration in Python applications

In September 2025, Google released the NotebookLM Enterprise API, enabling programmatic access to NotebookLM features for the first time.

While you can interact with the API using simple curl commands, this approach has several limitations that this project addresses:

• Authentication complexity

  • Problem: Managing OAuth tokens, handling token refresh, and ensuring secure credential storage
  • Solution: Seamless gcloud CLI integration with automatic token caching and refresh

• Manual request construction

  • Problem: Writing JSON payloads by hand, managing resource names, and handling API versioning
  • Solution: Type-safe CLI flags and Python SDK with intelligent defaults and validation

• Error handling

  • Problem: Cryptic HTTP error codes without context or recovery suggestions
  • Solution: Clear, actionable error messages with automatic retries for transient failures

• Repeated operations

  • Problem: Writing boilerplate loops for fetch/add/delete sequences
  • Solution: Higher-level client helpers and CLI flags that wrap single API calls (with retries built in) so scripts stay concise

• Output parsing

  • Problem: Manual JSON parsing and extracting specific fields from responses
  • Solution: Structured response objects in the Python SDK and --json output in the CLI for easy integration with tools like jq

This project provides production-ready tools that make the NotebookLM API accessible and reliable:

• Rust CLI: Fast, cross-platform binary for shell scripting and automation
• Python SDK: Idiomatic Python bindings for application integration
• Type safety: Compile-time checks prevent common API usage errors
• Developer experience: Intuitive commands and clear documentation

brew tap k-dash/nblm https://github.com/K-dash/homebrew-nblm
brew install k-dash/nblm/nblm
nblm --version

Download the tarball for your architecture from the Releases page, then extract and install it. Example for x86_64:

VERSION=x.x.x
ARCH=linux-x86_64   # or linux-aarch64
curl -LO https://github.com/K-dash/nblm-rs/releases/download/v${VERSION}/nblm-${ARCH}.tar.gz
tar -xzf nblm-${ARCH}.tar.gz
chmod +x nblm
sudo mv nblm /usr/local/bin/
nblm --version

Optional: verify the download with shasum -a 256 nblm-${ARCH}.tar.gz and compare with the digest listed on the release page.

git clone https://github.com/K-dash/nblm-rs.git
cd nblm-rs
cargo build --release -p nblm-cli
./target/release/nblm --version

You can also install the CLI locally with cargo install --path crates/nblm-cli.

pip install nblm
# or
uv add nblm

Prerequisite: a Google Cloud project with the NotebookLM Enterprise API enabled and either gcloud auth login or an OAuth token ready for NBLM_ACCESS_TOKEN.

For detailed setup and troubleshooting, see the documentation in docs/.

# 1. Authenticate
gcloud auth login

# 2. Set environment variables
export NBLM_PROJECT_NUMBER="123456789012"  # Get from GCP console
export NBLM_LOCATION="global"
export NBLM_ENDPOINT_LOCATION="global"

# 3. Create a notebook
nblm notebooks create --title "My Notebook"

# 4. Add a source
nblm sources add \
  --notebook-id YOUR_NOTEBOOK_ID \
  --web-url "https://example.com" \
  --web-name "Example"

from nblm import NblmClient, GcloudTokenProvider, WebSource

# Initialize client
client = NblmClient(
    token_provider=GcloudTokenProvider(),
    project_number="123456789012"
)

# Create a notebook
notebook = client.create_notebook(title="My Notebook")

# Add sources
response = client.add_sources(
    notebook_id=notebook.notebook_id,
    web_sources=[WebSource(url="https://example.com", name="Example")]
)

Platform	CLI	Python SDK
Linux
macOS
Windows

Complete guides and API references:

• Getting Started - Installation, authentication, configuration
• CLI Reference - All commands, options, and examples
• Python SDK Reference - API reference and usage patterns

• NotebookLM API Documentation - Official API documentation
• NotebookLM API Reference - API reference

Contributions are welcome! See CONTRIBUTING.md for development setup and guidelines.

MIT