Superset is a source-available agentic IDE with more than 15,400 stars on GitHub, built to orchestrate over a hundred AI coding agents running in parallel on your codebase.
Claude Code, Codex, and OpenCode are agents; Superset is where you run many of them at once. Each task gets its own isolated Git worktree and branch, so ten agents can work on ten branches simultaneously without overwriting each other’s files, while you monitor, review, and merge from a single window. A status view shows at a glance which agents are working, which are blocked, and which are waiting on you. Superset is agent-independent by design: pick a different CLI coding agent for every task, keep using your existing subscriptions and API keys, and never have model calls proxied or your workflow locked to one provider. Automations can start agents on events or schedules — issue triage, changelog drafts, dependency bumps — and open pull requests for your review. Workspaces run on any machine you add as a host, keep running when your laptop sleeps, and can be driven from the CLI, a TypeScript SDK, MCP, or the companion iPhone app. The desktop app runs on macOS with an experimental Linux build.
Superset has a free tier, with paid seats for teams that need more. The full source code is available on GitHub under the Elastic License 2.0, so you can inspect, audit, and self-host it subject to the license terms; you always use your own agent subscriptions and API keys.
Developers and engineering teams who already use AI coding agents and want to scale from one agent in a terminal to a coordinated fleet — without giving up their tools, accounts, or review habits.