inbox

Send, receive, and log atomic JSON messages between agents via CLI.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/MantisWare/BizForge --skill inbox-mantisware
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: inbox
Source: https://github.com/MantisWare/BizForge/tree/main/library/skills/coordination/inbox
Command: npx skills add https://github.com/MantisWare/BizForge --skill inbox-mantisware

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The communication backbone for multi-agent coordination. Every agent has an inbox directory. Messages are JSON files written atomically (write to temp, then rename — no partial reads). Supports direct messages, broadcasts, and reply chains. Full event log for debugging coordination issues.

Core Features & Use Cases

  • Point-to-point messaging between agents with per-agent inboxes.
  • Atomic file-based delivery with a reliable event log for traceability.
  • Supports direct messages, broadcasts, replies, and simple inbox inspection to coordinate workflows.

Quick Start

Send a direct message to an agent using the inbox command to begin a conversation.

Frequently Asked Questions about inbox

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

FAQPage Schema
How do I coordinate multiple agents with file-based messaging?

Multi-agent communication is handled by providing each agent with a dedicated inbox directory where messages are written as JSON files. The system supports direct messaging, broadcasts, and threaded replies through a simple CLI to coordinate workflows.

How does atomic file writing prevent partial reads in inter-agent communication?

Atomic file writing prevents partial reads by writing messages to a temporary file first, then renaming it to the final filename only after the write is complete. This guarantees agents never read incomplete JSON data during coordination.

Can I broadcast a message to multiple agents and trace message history?

Yes, you can broadcast messages to multiple agents simultaneously and trace message history through a unified event log. The CLI supports broadcast and log subcommands to inspect and debug multi-agent coordination issues across projects.

What is the best way to debug multi-agent coordination issues?

The best way to debug multi-agent coordination issues is by inspecting the unified event log that records all messaging activity. You can use the log subcommand with optional filters to trace message history and pinpoint communication failures across agents.

Do I need a database to enable messaging between agents?

No, you do not need a database to enable messaging between agents. The system relies entirely on a file-based inbox directory structure with atomic writes, making it a lightweight solution for multi-agent workflows without external dependencies.