Download for macOS
Skill

marimo-pair

@marimo-team Updated 2026-10-26

Work inside the user's live marimo notebook from the code editor: run Python in the same kernel the user does, inspect live notebook state, and commit durable notebook changes through code mode. Use whenever you create, analyze, or improve the user's marimo notebook.

artificial-intelligencedagdata-sciencedata-visualizationdataflowdeveloper-toolsmachine-learningnotebooks

Install

With your agent

Install OMGSkills skill "marimo-team/marimo:marimo-pair" into this agent's normal skills directory: use the OMGSkills MCP get_skill tool, install only the pinned_install commit and path, verify SKILL.md against pinned_install.skill_md_sha, and stop if MCP or the pin is unavailable. Do not overwrite or change existing skills without asking me.

Manual command

git clone https://github.com/marimo-team/marimo /tmp/marimo && ln -s /tmp/marimo/marimo/_server/ai/skills/marimo-pair ~/.claude/skills/marimo-pair

From README

Running Code in the Kernel You are already attached to the user's running notebook — there is nothing to connect to or start. Use the executecode tool to run Python in that kernel, passing your Python as the code argument. Everything you do — inspecting state, testing transformations, and persisting changes via cm — runs through executecode in the scratchpad (see below). Required First Kernel Command Start every code-mode session with this dedicated executecode call: Follow this order for every live kernel, including read-only tasks: Run only the inspection command above. Wait for successful help(cm) output. Use cm.getcontext() or another cm API in a later executecode call. Do not combine the inspection with task-specific code, and do not use another cm API before the inspection succeeds. This verifies the private, unstable API exposed by the marimo version in the user's active kernel.