mco-org/mco

What it solves

MCO addresses the problem of relying on a single AI coding agent's perspective, which can lead to blind spots in code review, implementation, or architecture analysis. It allows users to orchestrate multiple AI agents in parallel to compare their raw answers and make more confident decisions.

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 given, MCO dispatches the prompt to the selected agents, runs them in parallel, and collects their raw answers, operational status, and artifacts without automatically synthesizing them into a final decision.

Who it’s for

Developers who use AI coding agents and want to cross-reference multiple models for high-risk tasks, as well as other coding agents that can use MCO's self-describing CLI to coordinate multiple providers.

Highlights

• Multi-Agent Orchestration: Run multiple agents in parallel, chain their work, or divide the scope of a repository across different agents.
• Broad Provider Support: Integrates with a wide range of tools including Claude Code, Codex, Gemini, OpenCode, Qwen Code, GitHub Copilot, Hermes, Pi, Grok, and Cursor.
• Flexible Execution Modes: Supports read_only for reviews, write for workspace mutations, and yolo for broad bypass profiles.
• Agent-to-Agent Integration: Can be installed as a "Skill" for other coding agents, allowing them to trigger MCO workflows.