colony

Coordinate coding agents with claims, handoffs, and health monitoring.

3|Updated Apr 23, 2026

Search Tags:#software development #coordination #agent management #local-first #coding agents

Ready to run

npx skills add https://github.com/opencue/colony --skill colony

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: colony
Source: https://github.com/opencue/colony/tree/main/skills/build
Command: npx skills add https://github.com/opencue/colony --skill colony

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires colonyq, and includes scripts (resource) components.

What problem does it solve?

This Skill simplifies the coordination and management of coding agents, enhancing productivity and efficiency in software development workflows.

Core Features & Use Cases

Local-first Coordination: Ensures seamless collaboration among coding agents without the need for centralized services.
Claims and Handoffs: Manages tasks and resources, ensuring that agents work on the right tasks at the right time.
Health and Memory: Monitors the status and progress of agents, maintaining a history of interactions for better decision-making.
Use Case: Imagine you have a team of developers working on a complex project. Use this Skill to coordinate their efforts, track progress, and ensure that each developer is working on the right tasks without conflicts.

Quick Start

Install the colonyq package globally and initialize your colony environment. colony install --ide codex npx skills add recodeee/colony/skills/colony-mcp colony health

Frequently Asked Questions about colony

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I coordinate multiple coding agents like Claude Code and Codex on a single project?▼

Local-first coordination of coding agents like Claude Code, Codex, and Cursor is managed through claims and handoffs. This ensures each agent works on the right task at the right time without conflicts, tracking progress via health monitoring.

What's the best way to prevent task conflicts when running multiple AI coding agents locally?▼

Preventing task conflicts in local-first environments requires a coordination system that manages claims and handoffs. By monitoring agent health and maintaining interaction history, the system ensures developers and agents avoid overlapping work.

Do I need colonyq installed to manage coding agents with this Skill?▼

Yes, colonyq is a required dependency. You must install the colonyq package globally and initialize your environment before setting up agent runtimes, monitoring health, and managing task handoffs.

Can I use this Skill to monitor the health and memory of Gemini CLI and OpenCode agents?▼

Yes, health and memory monitoring supports Gemini CLI, OpenCode, and other agent runtimes. The system tracks agent status and progress while maintaining a history of interactions for better decision-making.

How does local-first agent coordination work without centralized services?▼

Local-first agent coordination operates without centralized services by managing claims, handoffs, and health monitoring directly in your environment. This architecture enables seamless collaboration among coding agents while maintaining local control.

Which coding agent runtimes are supported for task handoffs and claims management?▼

Supported coding agent runtimes include Claude Code, Codex, Cursor, Gemini CLI, and OpenCode. The system coordinates their activities through claims and handoffs, ensuring agents work on assigned tasks without conflicts.