Guide
API
Examples
Guide
API
Examples

CLI Overview

The Tarko Agent CLI is a flexible framework built on top of the Agent Kernel (@tarko/agent). It provides a comprehensive command-line interface for deploying and running agents with ease, featuring built-in Web UI and powerful extensibility.

Installation

Install globally via npm:

npm install -g @tarko/agent-cli

Or use with npx for one-time execution:

npx @tarko/agent-cli run my-agent

Quick Start

# Start interactive Web UI (default)
tarko

# Run with built-in agents
tarko run agent-tars  # Agent TARS
tarko run omni-tars   # Omni-TARS
tarko run mcp-agent   # MCP Agent

# Run with custom agent
tarko run ./my-agent.js

# Start headless API server
tarko serve

# Headless mode with direct input
tarko run --headless --input "Analyze current directory structure"

# Pipeline input
echo "Summarize this code" | tarko run --headless

Core Concepts

Deployment Modes

  • Interactive Mode (tarko run) - Web UI for development and testing
  • Server Mode (tarko serve) - Headless API server for production
  • Scripting Mode (tarko run --headless) - Silent execution for automation

Built-in Agents

Tarko CLI includes several ready-to-use agents:

  • agent-tars - Advanced task automation and reasoning system
  • omni-tars - Multi-modal agent with comprehensive capabilities
  • mcp-agent - Model Context Protocol agent for tool integration

Configuration Flexibility

Supports multiple configuration formats with auto-discovery:

  • tarko.config.ts - TypeScript configuration
  • tarko.config.yaml - YAML configuration
  • tarko.config.json - JSON configuration
  • Environment variables and CLI arguments

Key Features

Tool & MCP Server Filtering

Filter available tools and MCP servers via configuration:

tarko --tool.include "file_*,web_*" --tool.exclude "dangerous_*"
tarko --mcpServer.include "filesystem" --mcpServer.exclude "experimental_*"

调试支持

通过 --debug 参数启用详细日志：

# 启用调试模式
tarko run --debug

# 使用环境变量
DEBUG=tarko:* tarko run

Use Cases

Development

  • Interactive agent development with Web UI
  • Real-time debugging and testing
  • Hot reload and development mode

Production

  • Headless API server deployment
  • Docker and Kubernetes integration
  • Load balancing and scaling

Automation

  • CI/CD pipeline integration
  • Batch processing and scripting
  • Silent execution with structured output

Testing

  • Direct LLM requests for debugging
  • Agent behavior validation
  • Performance testing and benchmarking

Next Steps

API Reference

For detailed TypeScript definitions: