Reactive Python
Keep data, computation, and controls in one notebook.
Web views for reactive Python notebooks
Turn reactive Python into reports, apps, and presentations. Work with a coding agent, and keep every view connected to your analysis.
Skip to content
Example: Quadratic programs
Teach the same optimization problem as a lecture, a reading page, or a hands-on lab. Each view reads from the same Python notebook.
Example views
Each example notebook backs several views: slide decks, reading pages, linked explorers, maps, monitors, and printable reports. Open a view to compare it with its siblings and the notebook behind them.
LectureQuadratic programsReact · Reveal.js
OverviewRio 2016 athletesPlain HTML
StoryEarthquake watchPlain HTML · Observable Plot
MonitorBuilding occupancyNotebook Kit · Observable Plot
ExplorerRio 2016 athletesSvelte · Mosaic
OperationsEarthquake watchReact · MapLibre
Model reviewBuilding occupancyReact · Recharts
ExplainerQuadratic programsPlain HTML
BriefingEarthquake watchReact · Reveal.js · D3
PDF reportBuilding occupancyReact · React PDF
LabQuadratic programsSvelte · D3
FieldRio 2016 athletesPlain HTML · Shower · Three.js
Browse the examples or run one locally.
Ask a coding agent for a view
Paste one request into a terminal agent such as Claude Code or Codex:
text
Run `uvx --with marimo-studio agent-plugins read marimo-studio` and create
a scrollytelling report and a slide deck explaining calculus basics.The instructions ship with the marimo-studio package. They tell the agent to start a notebook with Studio, write its cells, then create, build, and show each view beside the notebook.
Author with a coding agent or follow Getting started by hand.