skills.rest
jwilger
jwilgerCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

agent-coordination

Enforce one-message-then-wait and shutdown protocols for multi-agent workflows.

1|Updated Jul 1, 2026
Search Tags:#workflow#multi-agent#orchestration#coordination#event-driven#handoffs
One-click install
npx skills add https://github.com/jwilger/agent-skills --skill agent-coordination-jwilger
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-coordination
Source: https://github.com/jwilger/agent-skills/tree/main/skills/agent-coordination
Command: npx skills add https://github.com/jwilger/agent-skills --skill agent-coordination-jwilger

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Multi-agent coordination discipline that enforces one-message-then-wait, idle heartbeat handling, no polling loops, sequential agent spawning, and proper shutdown protocols to prevent message spam, idle interruptions, and unsafe handoffs.

Core Features & Use Cases

  • One Message Then Wait: ensure complete context is delivered in a single interaction and avoid follow-ups before a reply.
  • Idle Heartbeat Handling: treat idle signals as normal progress, not a trigger to intervene.
  • No Polling Loops: promote event-driven coordination and harness-native completion signals.
  • Sequential Agent Spawning: avoid race conditions when launching multiple agents.
  • Agent Lifecycle Management: define safe startup/shutdown flows and preserve in-flight work.
  • Reference Patterns: Claude Code coordination, fallback coordination, and anti-pattern guidance.

Quick Start

Enable agent-coordination in ensemble workflows to begin orchestrating multi-agent tasks.

Frequently Asked Questions about agent-coordination

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

FAQPage Schema
How do I prevent message spam when orchestrating multi-agent workflows?▼

Multi-agent workflows prevent message spam by enforcing a one-message-then-wait discipline, ensuring agents deliver complete context in a single interaction and avoid follow-ups before receiving a reply.

What is the best way to handle agent handoffs without causing race conditions?▼

Safe agent handoffs require sequential agent spawning to avoid race conditions, combined with proper startup and shutdown protocols that preserve in-flight work during multi-agent orchestration.

Why does my event-driven coordination loop interrupt idle agents unnecessarily?▼

Idle agents are interrupted when idle signals are mishandled; event-driven coordination must treat idle heartbeats as normal progress rather than triggers to intervene or poll.

How do I manage agent lifecycle shutdown without losing in-flight work?▼

Agent lifecycle management defines safe startup and shutdown flows that preserve in-flight work by enforcing proper shutdown protocols and preventing unsafe handoffs during multi-agent coordination.

Do I need polling loops to check completion status across agent teams?▼

No, multi-agent coordination promotes event-driven coordination using harness-native completion signals instead of polling loops, preventing idle interruptions and unsafe handoffs across agent teams.

Related Skills

Explore More Software Engineering Skills→
AesopScottAesopScott
1

define-agent

Convert role artifacts into agent briefs with job, authority, and boundaries.

Community
Advanced
grandinhgrandinh
2

handoff-receiver

Parse and execute Cursor handoffs using the normalized handoff schema with IMPLEMENT mode gating.

Community
Basic
jcolanojcolano
1

flow-nexus-swarm

Orchestrates cloud-based AI agent swarms and automates event-driven workflows.

Community
Advanced
plurigridplurigrid
60

tmux

Manage tmux sessions, windows, and panes across terminal workflows.

Official
Basic