CLI Agent Orchestrator (CAO) is an open-source agent orchestration harness maintained by AWS Labs (awslabs). It provides a structured CLI runtime environment to coordinate, run, and manage AI-powered CLI agents and external tool integrations within developer environments. In practical use, CAO enables developers to execute complex workflows by orchestrating command-line agent tasks, integrating Model Context Protocol (MCP) servers, and interfacing with local dev setups. It includes native MCP application packages (`cao_mcp_apps`) to bridge agent reasoning with local system tools and developer workspace resources. What sets CLI Agent Orchestrator apart from standalone interactive coding agents is its focus on orchestration and containerized reproducibility. Through devcontainer feature support (`.devcontainer/features/cao`), CAO can be bundled directly into development container definitions, ensuring standardized agent execution across teams without manual environment setup. CLI Agent Orchestrator is fully open-source under the Apache 2.0 license and is maintained on GitHub by AWS Labs.
Tags: agent harness, cli, coding agent, OpenSource, terminal
cao_mcp_apps) to expose standard tool definitions and resources to agents. - Devcontainer Feature Support — Provides preconfigured devcontainer features (.devcontainer/features/cao) for seamless container-based developer setups. - Automated Workflow Execution — Runs structured agent routines and command pipelines directly from the terminal. - Open-Source AWS Labs Ecosystem — Built under standard open-source AWS Labs governance with transparent release cycles and testing suites..devcontainer/features/cao feature definition. This ensures all team members have identical CLI agent runtimes and MCP tool connections ready upon opening Codespaces or VS Code containers.cao_mcp_apps to connect agents with local MCP servers. This enables agents to query local repositories, run tests, and invoke system tools through standardized protocols.CLI Agent Orchestrator is designed for backend engineers, platform teams, and DevOps practitioners who want a standardized, container-friendly orchestration harness to manage AI agents and MCP tools in terminal environments. It is less suited for developers seeking an out-of-the-box GUI coding assistant or standalone chat UI.
bash # Clone the repository git clone https://github.com/awslabs/cli-agent-orchestrator.git cd cli-agent-orchestrator # Alternatively, add to your devcontainer.json # "features": { # "ghcr.io/awslabs/cli-agent-orchestrator/cao:2": {} # }bash # Navigate to the workspace and verify installation cao --version # Configure MCP apps if using external tools cd cao_mcp_apps npm installbash # Start CAO in your target project directory cao run --prompt "Analyze repository test coverage and run failing unit tests"Part of CLI Coding Agents
Categories: AI Coding Agents, Code Assistants, Developer Tools