Get the FREE Ultimate OpenClaw Setup Guide →

superset

IDE for the AI Agents Era - Run an army of Claude Code, Codex, etc. on your machine

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio superset-sh-superset node path/to/server.js \
  --env ENV_EXAMPLE="Placeholder or describe required vars if any"

How to use

Superset is a terminal-centric platform designed to run and manage multiple CLI coding agents in parallel, isolate each task in its own workspace, and monitor all active agents from a single interface. The server enables you to orchestrate agents that can execute commands, manage their own working directories via worktrees, and provide built-in diff viewing for reviewing changes. You can leverage Superset to run various CLI tools and agents that operate in a terminal, such as code linters, formatters, compilers, or AI-assisted coding agents, all within separate, isolated environments. To use the server, start the Node-based server (as configured in this MCP setup) and connect via its UI or API to spawn and manage multiple agents, switch between workspaces, and observe agent status and outputs in real time.

How to install

Prerequisites:

  • Git must be installed
  • Node.js or Bun environment (as required by the server build) and access to the project repository
  • macOS recommended (Windows/Linux untested for this project)

Installation steps:

  1. Clone the repository: git clone https://github.com/superset-sh/superset.git cd superset

  2. Install dependencies (example using Bun as in the project): bun install

  3. Run the development server: bun run dev

  4. (Optional) Build desktop app or other components as needed: bun run build open apps/desktop/release

Note: The exact commands may vary depending on how you host or deploy the MCP server. If you are integrating this repository as an MCP server in a larger environment, ensure the entry point (server.js or equivalent) is correctly referenced in the mcp_config and that any required environment variables are provided.

Additional notes

Tips and common considerations:

  • Ensure Bun is installed and available in your PATH since the project examples use bun run and bun install.
  • If you encounter port or proxy issues in development, configure any required reverse proxy (e.g., Caddy) as outlined in the project docs.
  • Review environment variables in .env.example and populate .env as needed for local testing; consider SKIP_ENV_VALIDATION if you want to bypass validation temporarily.
  • The MCP configuration shown uses a single server named 'superset'. You can expand by adding more servers under mcpServers with distinct command/arg configurations.
  • Monitor agent activity and workspace changes from the built-in UI, leveraging worktree isolation to prevent cross-task interference.

Related MCP Servers

Sponsor this space

Reach thousands of developers