Jupyter Introduces Its First Visual Debugger Plugin and xeus‑python Kernel Support

Jupyter has released its first visual debugging plugin and a supporting kernel, enabling breakpoint setting, variable inspection, and execution control directly within JupyterLab.

While Jupyter is traditionally used for interactive data analysis, the lack of debugging capabilities limited its use for larger codebases; the new plugin addresses this gap.

To install the front‑end extension, run:

jupyter labextension install @jupyterlab/debugger

The backend requires a kernel that implements the Jupyter Debug Protocol; currently xeus‑python is supported and can be installed with:

conda install xeus-python -c conda-forge

After installing both components, users can launch the visual debugger, try an online demo, and benefit from features such as a side‑panel file explorer, inline breakpoint setting, visual markers, and variable browsers with tree and table views.

The debugger is built on the Debug Adapter Protocol, making it compatible with any language kernel that supports debugging, and future improvements include richer variable rendering and conditional breakpoints.

The article also mentions a VS Code visual debugging tool that visualizes data structures in real time, offering alternative ways to inspect program state.