Claude Squad

Claude Squad is for the terminal person with more tickets than hands. You type cs, open a session per task, and each of Claude Code, Codex, Gemini CLI or Aider gets its own pane and branch while you watch the diff. The trait that decides it is that you never leave the shell; there is no window, no board, no account.

You need enough parallel work to justify it, three or four independent changes at once. Pick it if that is your week. Pick Conductor if you want a Mac app with review built in, and Agent Orchestrator if you want a board.

The failure mode is --autoyes. The README marks it experimental and it does exactly what it says: every permission prompt from every agent is answered yes, in parallel, on a real filesystem, because there is no container layer. One agent that decides to reinstall dependencies across a shared cache is four agents doing it. The other documented gotcha is startup timeouts, whose fix is to update the agent underneath.

Leave -y off and read each pane. What it does right: it adds no model layer of its own, so there is nothing of its own to hallucinate, only terminals to manage.

The mechanism is two Unix primitives. 1. Each agent runs in its own tmux session, so output and input are separated per task. 2. Each session gets its own git worktree on its own branch, so two agents editing the same file produce two commits rather than one corrupted tree. 3. A preview tab shows the diff; a keystroke commits and pushes.

There is no merge step: branches are pushed and reconciled elsewhere, by you. No benchmark is published and none is claimed. The observation: this is isolation without integration, which is the easier half of orchestration, done cleanly.

smtg-ai is a repository, not a company. There is no pricing, no funding announcement and no hosted anything, and at 8,422 stars it is one of the smaller projects here despite a March 2025 release that predates most of the category. The idea, one worktree per agent, has since been copied by every desktop orchestrator on this board with a cap table.

Moat: none; switching cost is deleting a binary. Likely outcome: not acquisition but absorption, the features living on in better-funded tools. Position: use it freely, expect no roadmap.

The demo is four panes and a diff. For sixty engineers the cost is zero on the tool and whatever the agents underneath already cost, so the budget line does not move. Procurement has nothing to sign and nothing to audit: no vendor, no SSO, no central log, and the prerequisites are tmux and the GitHub CLI on each machine.

It does not run in CI, so it changes nothing about our pipeline. Onboarding is a brew install for anyone already living in a terminal. Approved with conditions: personal use only, and the agents it wraps must already be on our approved list.

Go under AGPL-3.0, which is the copyleft variant, so if I bolt it into a hosted service I owe the source; for a laptop tool that costs me nothing. Configuration is ~/.claude-squad/config.json, and profiles let me define several named programs, so one profile can be Claude Code and another can be Aider pointed at my local Ollama, switched from a picker when the session starts.

There is no MCP and no model setting because it has no model; it inherits both from the agent in the pane. Forking is a go build. Small, readable, mine.