Lucien Working dir + jupyter Guidance

Lucien
1

Lucien Workspace Execution Environment

Lucien provides a robust Workspace Execution Environment that enables seamless integration of its intelligent analysis capabilities with your local files and codebase. With minimal configuration, you can authorize Lucien to access a designated local working directory, thereby unlocking advanced features such as local file operations, collaborative notebook editing, and direct access to a full-featured JupyterLab interface.

image
image1564×348 22 KB

Configuring the Local Working Directory

Once a local directory is successfully set as the working directory within Lucien, the platform gains the ability to interact with your local filesystem, enhancing its analytical and operational capabilities.

Local Filesystem Integration

Lucien’s local access empowers it with a range of file-level operations, including:

  • Advanced File Parsing: Lucien can read and extract content from both standard text files and complex structured documents, including .pdf, .docx, .pptx, and .xlsx formats.

  • Comprehensive Directory Operations: Lucien can recursively list entire directory structures and supports batch operations for reading multiple files at once.

  • Notebook Reading: Lucien can read the content of Jupyter Notebook (.ipynb) files for analysis or display.

  • Basic File Management: Lucien can create new files, update existing ones, and revert changes to notebook and other files.

Jupyter Integration

Lucien integrates tightly with a local Jupyter server, forming the foundation of its notebook collaboration capabilities. By enabling the “Enable code execution” toggle, Lucien enters a fully interactive mode that supports dynamic computation and notebook manipulation.

When Execution is Enabled

with toggle on
with toggle on1474×322 27.2 KB

Enabling the toggle grants Lucien full access to the local Jupyter server, unlocking the following capabilities:

  • Collaborative Notebook Editing: Lucien can add, update, and delete cells within .ipynb notebooks in real time.

  • Code and Shell Execution: Lucien can execute code within notebook cells and isolated code blocks. It can also run shell commands via a secure, sandboxed terminal, supporting workflows such as data analysis, code evaluation, and environment setup.

When Execution is Disabled

with toggle off
with toggle off1458×304 24.9 KB

Disabling the toggle restricts Lucien’s capabilities to:

  • Basic File Management: Lucien can still create, update, and revert changes to non-notebook files.

  • No Execution or Notebook Modification: Lucien is explicitly prohibited from executing code or modifying notebook files (.ipynb) by adding, deleting, or updating cells.

Accessing JupyterLab

Upon successfully configuring the local working directory, the “Open JupyterLab” button becomes available. Clicking this button launches a dedicated JupyterLab window, rooted at the selected local directory.

This interface is ideal for advanced development and debugging tasks such as:

  • Interactive code exploration and debugging
  • Managing Jupyter kernels, extensions, and terminals
  • Performing file operations via GUI
  • Editing multiple files in parallel

Registering a Custom Jupyter Kernel for Lucien

Lucien supports kernels from both uv environments and conda environments. Follow the steps below to register your own Python or R kernel.

Python Kernel

Option 1: Using uv

# Activate environment
uv run python

# Install ipykernel inside this env
uv pip install ipykernel

# Register kernel with Jupyter
python -m ipykernel install --user --name=myenv --display-name "Python (myenv)"

Option 2: Using conda

# Activate environment
conda activate myenv

# Install ipykernel
pip install ipykernel

# Register kernel with Jupyter
python -m ipykernel install --user --name=myenv --display-name "Python (myenv)"

R Kernel

  1. Start an R session inside your environment (uv or conda).

  2. Install IRkernel:

    install.packages("IRkernel", repos = "https://cloud.r-project.org")

  3. Register the kernel with Jupyter:

    IRkernel::installspec(name = "ir", displayname = "R (Standalone)")

Verify Kernel Installation

To confirm your kernel has been registered:

jupyter kernelspec list

You should see your custom environment (myenv) and/or R kernel in the list.