OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Multi-Agent Systems

Squads CLI

Squads CLI is an open-source TypeScript command-line tool for running teams of AI agents from a local Git repository. It defines role-based squads, stores goals, feedback, configuration and persistent memory in Markdown files, and invokes native AI command-line tools for conversational, directed and bounded runs.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Squads CLI — OpenViST project profile in Multi-Agent Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a task benefits from several specialized AI roles, repeated feedback cycles and a persistent local record of goals and outputs.

Who it is for

  • Developers building multi-agent workflows
  • Teams that want local, Git-backed agent state
  • Operators coordinating research, engineering, marketing, finance or operations tasks
  • Builders who need different AI CLI providers assigned to different agent roles

Documented uses

  • Coordinate lead, worker and verifier roles around a shared goal
  • Carry context, feedback and learnings between agent runs
  • Route squad roles to different supported AI command-line providers
  • Surface decisions awaiting human approval, rejection or deferral
  • Keep agent configuration and work state in inspectable Markdown files
Before you choose

Requirements & limits

The documented local setup requires Node.js 20 or later, Git and Claude Code. Claude Code is the default provider and must be installed and logged in before the first run; other providers are optional and require their own setup and credentials. The project is released under the MIT license.

  • Local execution keeps the machine, API keys and data in the documented local workflow.
  • Git-backed Markdown provides a human-readable record instead of a database or server state layer.
  • Role definitions and feedback cycles support distinct planning, execution and verification responsibilities.
  • Mixed-model routing is available through native AI command-line tools, with provider-specific setup still required.
  • The inbox provides a defined place for decisions that remain with a human operator.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Initialize a local AI agent squad workspace

Prepare a Git-backed workspace and verify the documented Squads CLI setup.

Before you start

  • Node.js 20 or later
  • Git
  • Claude Code installed and logged in
  • A local project directory

Steps

  1. Install squads-cli globally with npm.

  2. Create a project directory and initialize a Git repository.

  3. Run the initialization command to scaffold the agents directory and starter squads.

  4. Use the setup-checking command to identify missing local requirements.

  5. Run the documented demo command to verify the setup end to end.

Expected outcome

A local repository contains the starter agent structure and the documented setup check has been run.

Guide limitations

  • The setup depends on Claude Code and its login state when using the default provider.
  • Other providers require separate provider setup and credentials.
  • This recipe documents the repository's setup flow; it does not independently verify execution.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
4 October 2026
Original website title
agents-squads/squads-cli · Official homepage title · retained in its source language
How we review sourcesSuggest a correction