mco-org/mco

CLI-first orchestration for AI coding agents: run selected agents and models in parallel, compare raw answers, and coordinate review or implementation workflows.

What it solves

MCO provides a way to orchestrate multiple AI coding agents simultaneously. Instead of relying on a single model's perspective, it allows users to run the same task across different agents in parallel, helping to identify blind spots, compare raw answers, and increase confidence in the resulting code or analysis.

How it works

MCO acts as a lightweight CLI-first orchestration layer. It uses a shared adapter contract to interface with various provider CLIs (such as Claude Code, Codex, Gemini, and Copilot). When a task is dispatched, MCO runs the selected agents in parallel or in a chain, collects their raw answers or file-backed artifacts, and presents them to the user without modifying the natural-language output into a consensus decision.

Who it’s for

It is designed for developers who use AI coding agents for code review, implementation, architecture analysis, and CI checks, as well as other coding agents that can use MCO as a skill to coordinate multiple models.

Highlights

  • Multi-Agent Orchestration: Run tasks across multiple providers (e.g., Claude, Codex, Pi) in parallel or chained sequences.
  • Flexible Execution Modes: Supports read_only for reviews and write for workspace mutations.
  • Raw Answer Comparison: Retains complete raw answers from each invocation to avoid automatic, potentially biased synthesis.
  • Broad Provider Support: Integrates with a wide range of coding agent CLIs including Gemini, Qwen, Grok, and Cursor.
  • Agent-to-Agent Integration: Can be installed as a "Skill" for other coding agents to call.

Related

  • Project
    awslabs/cli-agent-orchestrator

    A coordination framework that manages multiple AI coding CLIs, allowing a supervisor agent to delegate tasks to specialist workers in parallel or sequence.

  • Project
    stablyai/orca

    An AI orchestrator that lets developers run multiple CLI coding agents in parallel isolated worktrees to compare and merge their results.

  • Project
    code-yeongyu/oh-my-openagent

    A genuine AI‑agent toolkit that runs as an OpenCode plugin, a Codex‑CLI component set, or a standalone beta command. It ships 11 pre‑configured agents, a one‑word `ultrawork` launch command, Team Mode for parallel agents, LSP/AST‑grep integration, and optional hash‑line editing. Install via `bunx oh-my-openagent install`, `npx lazycodex-ai install`, or `npm i -g omo-ai@beta`.

  • Project
    YoanWai/agent-manager

    A terminal-based manager for AI coding agents that consolidates multiple CLI agents into a single interface with live status tracking and integrated diff reviews.