Skip to main content

Overview

⚠️Warning

Preview Release - macOS and Linux Only

Caret CLI is currently in preview and only available for macOS and Linux users. Windows support is coming soon.

ℹ️Note

이 매뉴얼은 Caret 기반으로 빌드된 캐럿 CLI를 기준으로 합니다. 기능·옵션은 Caret v3.38.1 머지본을 따르며, 캐럿 브랜딩/UI 차이는 최소화되어 있습니다.

What is Caret CLI?​

Caret CLI runs AI coding agents directly in your terminal. Pipe git diffs for automated code reviews in CI/CD, run multiple instances simultaneously for parallel development, or integrate Caret into your existing shell workflows.

The CLI tracks instances across your system and outputs in formats designed for both humans and scripts—JSON, plain text, or rich terminal output.

💡Tip

Ready to get started? Check out the installation guide to install Caret CLI and run your first task.

Supported Model Providers​

Caret CLI supports multiple AI model providers, giving you flexibility in choosing the best model for your needs:

  • Anthropic
  • OpenAI
  • OpenAI Compatible
  • OpenRouter
  • X AI (Grok)
  • AWS Bedrock
  • Google Gemini
  • Ollama
  • Cerebras

During installation, you'll authenticate and configure your preferred provider using the caret auth command.

What you can build with this​

Automated code maintenance

  • Schedule daily runs to identify and fix linting issues across your codebase
  • Create tasks that scan for security vulnerabilities and automatically patch them
  • Build scripts that update deprecated dependencies and run tests

Multi-instance development

  • Run separate Caret instances for frontend and backend simultaneously
  • Spawn instances for different feature branches, each with isolated state
  • Create parallel review processes for multiple PRs

Custom workflows

  • Build shell scripts that combine Caret with git hooks for pre-commit analysis
  • Create custom commands that pipe complex data structures through Caret for processing
  • Integrate with your existing toolchain (jq, grep, awk) for sophisticated automation

CI/CD integration

  • Add Caret to GitHub Actions for automatic code review on every PR
  • Create GitLab pipelines that generate migration scripts from schema changes
  • Build Jenkins jobs that use Caret to analyze test failures and suggest fixes

Learn more​

Installation

Install Caret CLI and authenticate with your account to get started.

Three Core Flows

Master the three ways to use Caret CLI: interactive mode, headless automation, and multi-instance parallelization.