/marimo-pair

reactive Python notebooks as environments for agents

Prerequisites

• A running marimo notebook (--no-token for auto-discovery; MARIMO_TOKEN env var for servers with auth)
• bash, curl, and jq available on PATH (on Windows, run from Git Bash, or from WSL with those tools installed in the distro — the scripts also find notebooks running on the Windows host)

Install

Agent Skills (any tool)

Works with any agent that supports the Agent Skills open standard:

npx skills add marimo-team/marimo-pair

# or upgrade an existing install
npx skills upgrade marimo-team/marimo-pair

If you don't have npx installed but have uv:

uvx deno -A npm:skills add marimo-team/marimo-pair

Claude Code (plugin)

Add the marketplace and install the plugin:

/plugin marketplace add marimo-team/marimo-pair
/plugin install marimo-pair@marimo-pair

To opt in to auto-updates (recommended), so you always get the latest version:

/plugin → Marketplaces → marimo-pair → Enable auto-update

FAQ

I keep getting prompted to allow Bash commands

The skill declares its own allowed-tools, but Claude Code may still prompt you to approve each Bash call. To avoid repeated prompts, copy the absolute paths to the scripts from the installed skill and add them to your .claude/settings.json (project-level) or ~/.claude/settings.json (global):

{
  "permissions": {
    "allow": [
      "Bash(bash /path/to/skills/marimo-pair/scripts/discover-servers.sh *)",
      "Bash(bash /path/to/skills/marimo-pair/scripts/execute-code.sh *)"
    ]
  }
}

Contributing

We'd love your help in improving this skill. After a marimo pair session, please run the /retro-marimo-pair command to share feedback on how we can improve the skill. You can also file an issue on the GitHub repository.