Skip to documentation
Documentation
Documentation
OverviewInstallationQuickstartConfigurationTroubleshootingFAQ

Usage

Terminal interfaceCLIDashboardAnalytics

Collaborate

ArtifactsSession sharingBlueprintsAgent identitiesMCP server

Configure

AgentsSkillsRulesCommandsInheritancePlaybooksExperimentsHooksTeamWorkspaces

API Reference

API OverviewAuthenticationSessions APIResources APIOrganizations APIProjects APIPlaybooks APIExperiments API
Need a hand?Find an answer in troubleshooting.
Loading docs…
CREWKIT / DOCUMENTATION

Installation

Install the crewkit CLI on macOS, Linux, or Windows via WSL.

Install the crewkit CLI to start managing your team's AI coding sessions.

brew install karibew/tap/crewkit

Verify the installation:

crewkit --version

Prerequisites

  • Claude Code (claude on your PATH) — required only for crewkit code, which launches an observed session. crewkit auth login, crewkit init and crewkit mcp serve do not need it.
  • Git — crewkit detects your project from the git remote
  • Node.js 18+ (if installing via npm)

Platform notes

macOS — Homebrew is the recommended install method. Apple Silicon (arm64) is supported; Intel (x64) builds are not yet available.

Linux — Use the curl installer or npm. Builds are available for x64.

Windows — Use WSL and install the Linux build with any command above. Native Windows builds are paused, so the Chocolatey package is frozen at 0.4.1 and npm no longer ships a Windows binary. WSL also enables background observation, which is Unix-only.

Updating

crewkit update

To check for updates without installing:

crewkit update --check

For script-installed binaries, crewkit also checks for updates in the background (at most hourly) and installs them automatically after verifying the release signature; set CREWKIT_NO_AUTO_UPDATE=1 to disable. Installs owned by Homebrew, npm, or Chocolatey are never touched — crewkit tells you the right package-manager command instead.

Next steps

  • Quickstart: your first session
  • Configuration

Related

Once the CLI is connected, the marketplace is where you pick up a ready-made agent, skill or playbook instead of writing one from scratch.

Overview

CrewKit is the institutional memory and governance layer for AI-assisted engineering: shared context, team rules and session visibility for every agent run.

Quickstart

Go from install to your first tracked session in under a minute.

On this page

PrerequisitesPlatform notesUpdatingNext steps