Question 1

What is AgentMux?

Accepted Answer

AgentMux is an open-source agent operating environment. Every agent runs as a first-class pane with its own identity, memory, and structured UI. Agents can drive the workspace itself via the Agent App API — opening panes, building dashboards, and messaging peer agents.

Question 2

Is AgentMux free?

Accepted Answer

Yes. AgentMux is free and open-source, licensed under Apache 2.0. You can download it from GitHub and use it without any cost.

Question 3

What platforms does AgentMux support?

Accepted Answer

AgentMux runs on Windows (x64), macOS (Apple Silicon), and Linux (x86_64). It bundles Chromium via CEF, so the UI renders the same on every platform.

Question 4

What programming language is AgentMux built with?

Accepted Answer

AgentMux has a 100% Rust async backend powered by Tokio and Axum, with a TypeScript/SolidJS frontend rendered in Chromium via CEF.

Question 5

How is AgentMux different from tmux?

Accepted Answer

Unlike tmux, AgentMux is a GUI desktop application with native AI agent integration, web views and real-time system monitoring. It is designed specifically for AI-assisted development workflows.

Question 6

How much memory does AgentMux use?

Accepted Answer

Memory use depends on how many panes and agents you have open. AgentMux bundles Chromium via CEF for the UI and runs a pure Rust backend with no Node.js runtime and no garbage collector; agent CLIs run as separate processes and use their own memory. The v0.57.6 download is about 142-198 MB depending on the format, and the portable builds take about 437-443 MB on disk.

Question 7

What is the Agent App API?

Accepted Answer

The Agent App API lets AI agents create panes and modify the workspace UI programmatically. Agents can spawn terminals, editors, and webviews, rename panes and tabs, open and close panes, open and switch tabs, and show custom dashboard views in a webview pane from within agent code.

Question 8

What is interpane reactive communication?

Accepted Answer

Interpane reactive communication lets agents message each other in real time. With SendMessage, one agent's message lands directly in another agent's input, enabling reactive pipelines where agents respond to each other automatically. Every message shows up in both agents' panes, tagged with its sender and trust level.

Question 9

Can AgentMux be used for legal work?

Accepted Answer

Yes. Legal teams use AgentMux to run parallel research agents that review case law, analyze contracts, and check regulatory compliance. Agents cross-verify each other's findings through agent-to-agent messaging, and each agent's session is kept locally for review.

Support

Report a bug or request a feature

Ask a question / join the community

Privacy or account questions