Question 1

What is an Agent Notebook?

Accepted Answer

A notebook where you and a coding agent work on the same analysis. In SpurLab, a connected ACP agent can use notebook context to help write cells, explore data, and explain results. The analysis remains in Python, SQL, charts, and notes you can inspect.

Question 2

Can I use my existing coding agent?

Accepted Answer

Yes, if it supports the Agent Client Protocol (ACP), directly or through a compatible integration. Use the agent’s own installation and authentication setup. SpurLab gives it a notebook to work in.

Question 3

Can I use my Jupyter notebooks?

Accepted Answer

Yes. Open .ipynb notebooks and work with Python cells, outputs, and Markdown. SQL cells let you explore data with DuckDB alongside your Python work.

Question 4

Does everything run locally?

Accepted Answer

The desktop app, local kernels, and notebook files live on your machine. A connected agent may send notebook context to its AI provider; that depends on the agent and provider you use. Connected APIs also use their respective services.

Question 5

Can I still work without an agent?

Accepted Answer

Yes. Open a notebook and work directly with Python and SQL. You can write, edit, and run cells yourself, and bring in an agent when you want help.

Question 6

Where should I start?

Accepted Answer

Install SpurLab, open a notebook, and connect your ACP-compatible coding agent. Start with a CSV you know. Ask the agent to inspect the columns, suggest an analysis, and write a cell you can check.

A fast Jupyter notebook. With your coding agent.