awslabs/cli-agent-orchestrator

AG-UI Author lets an agent emit declarative dashboard components through the emit_ui MCP tool. Use it to show approvals, progress, choices, metrics, and diff summaries in a live AG-UI stream.

AI-DLC Portfolio coordinates multiple AI-DLC workflows across repositories and worktrees with a catalog of evidence-backed projects, dependencies, intents, and dispatch state. Use it to initialize a portfolio workspace, register work, create worktrees, validate readiness, and monitor parallel sessions.

Add UI To MCP Server helps you turn existing MCP tools into interactive UI-backed app tools using MCP Apps. Use it when adding views to a server that already exposes tools and should keep text-only support.

CAO Agent Routing finds installed CAO agent profiles that match a task and selects the best one before assignment. Use it when routing work by capability instead of guessing a profile name.

CAO MCP Apps enables and extends CAO’s host-rendered fleet UI inside MCP App hosts like Claude Desktop, ChatGPT, VS Code Copilot, Goose, and Postman. Use it to turn on the surface, rebuild bundles, or add and troubleshoot ui://cao views.

CAO Memory stores, recalls, and forgets durable facts across sessions and providers. Use it to save preferences, project conventions, corrections, and other details the user should not have to repeat.

CAO Plugin Creator shows how to build a CAO plugin that handles lifecycle and messaging events. Use it to scaffold plugin packages, discover available hooks, and connect CAO to external systems like Slack, Discord, metrics, or dashboards.

CAO Provider Creator explains how to add a new CLI agent provider to CLI Agent Orchestrator. Use it to scaffold adapters, wire provider registration, and understand CAO's provider architecture.

CAO Session Liveness verifies whether a CAO session is truly running by comparing terminal status with the latest output. Use it when a session may have exited, stalled, or been misreported as active.

CAO Session Management launches and manages named multi-agent sessions in CLI Agent Orchestrator. Use it to start a conductor-led workflow, send messages, check status, and shut sessions down.

CAO Worker Protocols defines how worker agents handle assigned versus handed-off tasks in CLI Agent Orchestrator. Use it to decide when to return results directly and when to call send_message with the right receiver.

Cao Learning reports task outcomes and stores durable lessons so future CAO sessions can reuse them. Use it when supervising or working in a CAO workflow that supports outcome logging and memory.

Cao Supervisor Protocols covers dispatching worker agents with assign and handoff, plus idle inbox delivery and prompt responses in CLI Agent Orchestrator. Use it when supervising parallel or blocking tasks through CAO.

Cao Workflow author and run CAO Python workflow scripts for multi-step, parameterized agent jobs with fan-out and resume support. Use it when a task needs validation and coordinated steps instead of a single agent call.

Convert Web App helps adapt an existing web application to run both standalone and as an MCP App. It is used when you need to analyze the app, map its inputs, and add server-side tool and resource registration.

Create MCP App explains how to build interactive UIs for MCP-enabled hosts by combining an MCP tool with an HTML resource. Use it when scaffolding an MCP App, adding UI to a tool, or wiring host integration and lifecycle patterns.

MCP Apps Builder loads the official ext-apps skills for creating, migrating, or extending MCP Apps with interactive ui:// views. Use it when building a new MCP App, adding a view to a server, converting an OpenAI Apps SDK app, or hybridizing a web app.

OpenAI App Migration helps convert OpenAI Apps SDK applications to the MCP Apps SDK. Use it when porting from window.openai, skybridge, or openai/outputTemplate, with mapping tables and migration steps.