Can I talk to agents other than Codex?

Yes, you can communicate with any agent configured in your environment, including Gemini or custom-named agents, using the 'talk' command.

How does the skill handle rate limiting?

The skill includes a --delay flag and a pasteEnterDelayMs configuration to control the speed of communication, helping to avoid rate limits and buffer overflows.

What is the primary purpose of the Tmux Multi-Agent skill?

It allows Claude to communicate with and control other AI agents running in separate tmux panes, enabling a collaborative multi-agent development environment.

How do I see which agents are currently available?

You can run the 'tmux-team list' command to view a complete list of all configured and active agents in your tmux session.

Is there a way to wait for an agent's response automatically?

Yes, by using the --wait flag with the 'talk' command, the skill will block until the target agent replies or a specified timeout is reached.

Tmux Multi-Agent Collaboration

Name: Tmux Multi-Agent Collaboration
Author: wkh237

bywkh237

•

Collaboration Tools

Enables seamless communication and coordination between multiple AI agents operating within distinct tmux panes.

About

This skill provides a specialized interface for Claude to orchestrate complex workflows by interacting with other AI agents (such as Codex or Gemini) running in parallel tmux environments. By leveraging the tmux-team CLI, it allows for targeted messaging, broadcast notifications, and synchronized request-response cycles, making it an essential tool for developers building multi-agent systems or requiring specialized model collaboration within a terminal environment.

Key Features

Configurable message delays to manage API rate limits
Broadcast messaging capabilities to all active agents
Synchronous 'wait' mode for blocking request-response patterns
Targeted messaging to specific AI agents (Codex, Gemini, etc.)
1 GitHub stars
Adjustable response retrieval for reading long agent outputs

Use Cases

Orchestrating a multi-agent pipeline for debugging, documenting, and testing code
Coordinating a code review across multiple specialized AI models simultaneously
Managing complex terminal workflows that require cross-model reasoning and data passing

About

Key Features

Configurable message delays to manage API rate limits
Broadcast messaging capabilities to all active agents
Synchronous 'wait' mode for blocking request-response patterns
Targeted messaging to specific AI agents (Codex, Gemini, etc.)
1 GitHub stars
Adjustable response retrieval for reading long agent outputs

Use Cases

Orchestrating a multi-agent pipeline for debugging, documenting, and testing code
Coordinating a code review across multiple specialized AI models simultaneously
Managing complex terminal workflows that require cross-model reasoning and data passing