Docs · Overview

How the manager works.

agent-manager is a Go TUI that runs on top of tmux. Every agent you start is a real tmux session, so the manager is a window onto them rather than a thing they live inside. Quit it and they keep working.

The session list: five agent sessions grouped under backend, frontend and infra, each with a colour-coded live status, a preview of the selected agent's pane on the right, and machine gauges at the bottom left. — Five sessions, three projects. One is blocked on a permission prompt.

The first run

There is no config to write and no daemon to start. Install it, run agent-manager, and a config file appears at ~/.config/agent-manager/config.toml on Linux or ~/Library/Application Support/agent-manager/config.toml on macOS, already holding working blocks for Claude Code, OpenCode, Codex and Grok Build.

n opens the new-session form (name, tool, directory, an optional starting prompt, and a group). g makes a group. ? shows every key without leaving the app.

Where the sessions live

Agent sessions run on a private tmux server named agentmgr, in the am_* namespace, so they never mix with the tmux you run yourself and a kill-server on your own socket leaves them alone. To reach one from a plain shell, name that server:

tmux -L agentmgr ls
tmux -L agentmgr attach -t am_<id>

Inside a managed session ctrl+q detaches back to the manager and ctrl+r flags a review on the way out. Both bindings are scoped to managed sessions, so they pass straight through everywhere else.

Where to go next

InstallHomebrew, the install script, the AUR, mise, Go, or a binary.
StatusWhy a row is orange, and where that verdict comes from.
Quick promptAnswering a blocked agent without attaching to it.
Diff reviewWhole-file diffs whose line comments go back to the agent.
ConfigurationTeaching it a CLI it has never heard of.