agent-coordination

Orchestrate Claude Code agents through registry-driven workflows with context injection and verification scripts.

80|15|Updated Nov 16, 2025
One-click install
npx skills add https://github.com/ilyasibrahim/claude-agents-coordination --skill agent-coordination-ilyasibrahim
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-coordination
Source: https://github.com/ilyasibrahim/claude-agents-coordination/tree/main/claude-user/skills/agent-coordination
Command: npx skills add https://github.com/ilyasibrahim/claude-agents-coordination --skill agent-coordination-ilyasibrahim

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Coordination and orchestration across multiple Claude Code agents, maintaining persistent memory and controlled handoffs to prevent context drift and task collisions.

Core Features & Use Cases

  • Dual registry management: _registry.md and _tech-debt.md to track completed work and deferred improvements.
  • Handoff protocols: context injection to ensure main agent governs prompts; subagents do not access registries directly.
  • Verification and archiving: provides verify.sh and archive script for maintaining and cleaning history.

Quick Start

Activate the main agent coordination flow to check prior work, inject context, sequence tasks, and verify deliverables using the included scripts and registries.

Frequently Asked Questions about agent-coordination

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

FAQPage Schema
How do I coordinate multiple Claude Code agents without losing context?

Agent coordination prevents context drift across multiple Claude Code agents by using a registry-driven workflow, enforcing handoff protocols where the main agent acts as the sole source of truth for prompt governance.

What is the best way to track deferred improvements when orchestrating agent workflows?

Agent coordination provides a dual registry system using _registry.md to track completed work and _tech-debt.md to log deferred improvements, ensuring persistent memory across orchestrated workflow tasks.

How do I sequence tasks and verify deliverables for automated agent workflows?

Agent coordination sequences tasks and verifies deliverables using included verification scripts like verify.sh, which check prior work against the registry before archiving completed items with explicit dependencies.

Can subagents access the registry directly during task handoffs?

No, agent coordination enforces handoff protocols where subagents do not access registries directly; the main agent injects context and governs prompts to maintain controlled handoffs and prevent task collisions.

Do I need any external dependencies to automate agent handoffs and archiving?

No external dependencies are required; agent coordination relies solely on internal scripts and markdown registries to automate workflow orchestration, context injection, verification, and the archiving of completed work.