message-bus

Coordinate distributed agents via a file-based message bus with append-only queues and event signals.

2|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/AhmedElhadarey/Gilfoyle --skill message-bus-ahmedelhadarey
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: message-bus
Source: https://github.com/AhmedElhadarey/Gilfoyle/tree/main/skills/message-bus
Command: npx skills add https://github.com/AhmedElhadarey/Gilfoyle --skill message-bus-ahmedelhadarey

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

File-based message bus enabling reliable inter-agent coordination among workers and board directors, with atomic writes, heartbeat tracking, and event-driven signals to prevent race conditions and improve auditability.

Core Features & Use Cases

  • Append-only message queue (queue.jsonl) for durable, ordered events.
  • Shared status and lock state (worker-status.json, locks.json) with simple expiry and safety checks.
  • Board deliberation artifacts (assessments.json, votes.json, discussion.jsonl) to support structured governance.
  • Use Case: orchestrating parallel tasks across dozens of agents while detecting deadlocks and coordinating decisions via signals.

Quick Start

Initialize the message bus for your track and begin coordinating agents using the built-in queue and event signals.

Frequently Asked Questions about message-bus

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

FAQPage Schema
How do I coordinate distributed agents without race conditions?

You coordinate distributed agents using a file-based message bus that applies atomic writes, heartbeat tracking, and event-driven signals to prevent race conditions. This ensures reliable inter-agent communication across a track's lifecycle.

What is a file-based message bus for inter-agent communication?

A file-based message bus is an append-only queue system using files like queue.jsonl to enable durable, ordered events. It supports safe, auditable communication across workers and directors using shared state files and event signals.

Can I detect deadlocks when orchestrating parallel tasks across many agents?

Yes, you can detect deadlocks during multi-agent orchestration by applying heartbeat tracking and lock state expiry checks. The system monitors worker-status.json and locks.json to identify stalled processes and coordinate decisions safely.

How do I structure board deliberations for multi-agent governance rounds?

You structure board deliberations using governance artifacts like assessments.json, votes.json, and discussion.jsonl. These files support structured governance rounds by capturing director decisions and preserving an auditable trail of the coordination process.

Does a file-based queue work for distributed systems orchestration at scale?

A file-based queue works for distributed systems orchestration across dozens of agents and governance rounds. It applies an append-only event queue with shared status files and safety checks to handle multi-worker coordination effectively.

What are the limitations of using file locks for distributed agent coordination?

File locks for agent coordination require simple expiry and safety checks to function reliably. If cleanup workflows are not properly managed, orphaned locks or stale heartbeats in worker-status.json may stall the coordination pipeline.