Marimo

## Editorial notes

Marimo is useful when Claude-adjacent teams need reproducible Python notebooks that can move between exploratory analysis, reviewable source code, scheduled scripts, SQL-backed data work, and deployable apps. Because notebooks are stored as pure Python, they are easier for agents and reviewers to diff, search, refactor, test, and version-control than hidden-state notebook formats.

This entry covers the open-source Marimo notebook project. It is distinct from Jupyter, Streamlit, Chainlit, DuckDB, and Polars. Jupyter is the traditional interactive notebook environment. Streamlit and Chainlit primarily focus on app or chat interfaces. DuckDB is an embedded analytical database. Polars is a DataFrame query engine. Marimo focuses on reactive, reproducible Python notebooks that can also run as scripts and apps.

## Source notes

- The official repository describes Marimo as a reactive Python notebook that runs reproducible experiments, queries with SQL, executes as a script, deploys as an app, versions with git, and stores notebooks as pure Python.
- The README highlights reactive execution, interactive UI elements, git-friendly Python files, SQL support, AI-native workflows, package management, script execution, app deployment, WASM support, reusable code, and testability.
- The README says Marimo can replace parts of Jupyter, Streamlit, Jupytext, ipywidgets, and Papermill workflows by combining notebooks, apps, scripts, and automation-oriented execution.
- The reactivity docs say Marimo builds a dependency graph from variable definitions and references, automatically marks dependent cells for execution, eliminates hidden state, and supports lazy runtime behavior for expensive notebooks.
- The reactivity docs also note that Marimo does not track variable mutations across cells and that global variable names must be unique.
- The app docs say `marimo run` starts a web server and runs a notebook as an app, with notebook code hidden by default in app view.
- The deployment docs distinguish edit servers for creating, running, and editing notebooks from app servers for read-only web apps, and cover public sharing, authentication, health endpoints, and traceback handling.
- The authentication docs describe token and password protection, note that there is no built-in authorization system, and suggest ASGI middleware for custom authentication.
- The SQL docs say Marimo SQL cells can query Python DataFrames, files, and databases, interpolate Python values, use the built-in SQL engine, and return results as DataFrames or DuckDB relations.
- The database docs under SQL describe connections for PostgreSQL, MySQL, SQLite, DuckDB, Snowflake, BigQuery, SQLAlchemy, SQLModel, and other database engines.
- The remote-storage docs describe browsing and using storage connections through `fsspec`, including S3, Google Cloud Storage, and Azure storage examples.
- The package-management docs and README describe serializing notebook requirements and auto-installing dependencies in isolated environments.
- The AI completion docs describe full-cell generation, inline completion, assistant context, referenced variable values, hosted providers, local Ollama models, GitHub Copilot, Windsurf, and custom copilot integrations.
- The LLM provider configuration docs cover OpenAI, Anthropic, Google AI, OpenAI-compatible endpoints, Ollama, Bedrock, Azure, OpenRouter, and other provider settings.
- The repository is `marimo-team/marimo`, is Apache-2.0 licensed, active, and maintained by the Marimo team.

## Duplicate check

Checked current `content/tools/`, `content/mcp/`, agents, hooks, rules, skills, commands, guides, collections, open pull requests, live issue state, and repository-wide content for `Marimo`, `marimo-team/marimo`, `github.com/marimo-team/marimo`, `docs.marimo.io`, `marimo.io`, `reactive notebooks`, `Python notebooks`, and `SQL cells`. No dedicated Marimo tools entry, source URL duplicate, target file, issue duplicate, semantic duplicate, or open duplicate PR was found.

## Disclosure

Editorial listing. No paid placement or affiliate link is used. Marimo is Apache-2.0 open-source software; Marimo Cloud or Molab, hosted AI providers, local model servers, GitHub Copilot, databases, warehouses, object stores, notebooks, cloud hosts, and downstream deployment services may have separate licenses, billing, terms, privacy obligations, and access controls.