agent-inbox

Facilitate asynchronous inter-agent communication through file-based JSON message queues.

1|Updated Apr 18, 2026
One-click install
npx skills add https://github.com/CC90210/CMO-Agent --skill agent-inbox-cc90210
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-inbox
Source: https://github.com/CC90210/CMO-Agent/tree/main/skills/agent-inbox
Command: npx skills add https://github.com/CC90210/CMO-Agent --skill agent-inbox-cc90210

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill solves the observability and communication gap in multi-agent systems where agents are forced to rely on synchronous, blocking calls, preventing efficient background task coordination.

Core Features & Use Cases

  • Async Delegation: Allows agents like Bravo, Maven, and Atlas to exchange messages without waiting for an immediate response.
  • Checkpoint-based Pickup: Enables agents to process incoming requests at their next natural activation point rather than polling.
  • Use Case: When a background task finishes, the worker agent posts a completion message to the inbox, which the orchestrator agent picks up during its next session start, ensuring seamless workflow continuity.

Quick Start

Use the agent-inbox skill to post a message from the codex agent to the bravo agent notifying it that the background task is complete.

Frequently Asked Questions about agent-inbox

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

FAQPage Schema
How do I handle asynchronous messaging between autonomous agents without blocking my orchestrator?

Asynchronous messaging between autonomous agents is handled through a file-based messaging protocol that eliminates blocking orchestrator dependencies. A priority-based queue system uses JSON files to ensure reliable task handoffs without waiting for immediate responses.

What is the best way to coordinate multi-agent workflows without synchronous polling?

Multi-agent workflow coordination without synchronous polling is achieved using checkpoint-based pickup. Agents process incoming requests at their next natural activation point, allowing a worker agent to post a completion message to the inbox for the orchestrator to pick up during its next session.

Can I use JSON files for priority-based task queues in a multi-agent system?

JSON files can be used to implement a priority-based queue system for multi-agent systems. This approach manages structured message passing between autonomous agents to ensure reliable task handoffs and status updates.

Why does my multi-agent orchestrator wait for background tasks to finish before continuing?

Multi-agent orchestrators wait for background tasks when using synchronous, blocking calls. Implementing an asynchronous file-based messaging protocol allows worker agents to post completion messages to an inbox, which the orchestrator picks up later to ensure seamless workflow continuity.

Does agent-inbox work for delegating background tasks to multiple autonomous agents?

Agent-inbox works for delegating background tasks to multiple autonomous agents by facilitating asynchronous inter-agent communication. Agents like Bravo, Maven, and Atlas can exchange messages without waiting for an immediate response, ensuring smooth background task coordination.

When do I need a file-based messaging protocol for multi-agent coordination?

A file-based messaging protocol for multi-agent coordination is needed when you have an observability and communication gap in your system. It prevents inefficient background task coordination caused by forced synchronous, blocking calls between agents.