agent-manager
Terminal multiplexer for AI coding agents. Run Claude Code, Codex, OpenCode, Gemini CLI and more side by side in one tmux TUI, with live status, worktrees and d
What is agent-manager?
Terminal multiplexer for AI coding agents. Run Claude Code, Codex, OpenCode, Gemini CLI and more side by side in one tmux TUI, with live status, worktrees and diff review. Discord: discord.gg/ecS88An7pF
Overview of agent-manager
What it is
agent-manager is a local terminal application that connects to existing AI coding agent subscriptions and tools. It runs multiple AI coding agents—such as Claude Code, Codex, OpenCode, Grok, Gemini CLI, Antigravity CLI, Pi, Command Code, Hermes Agent, Muse Code, Oh My Pi, and Cursor CLI—side by side, each in its own persistent tmux session. It is a thin layer over the CLIs you already have, carrying over your login, subscription, config files, MCP servers, and every feature the tool ships without modification. It runs on macOS, Linux, natively on Windows with psmux, and inside WSL2. The license is Apache-2.0 and the tool is open source.
Key features
- Each agent runs in its own persistent tmux session with live status displayed in one list, grouped into a project tree that can be folded and reordered.
- Send a prompt directly into any session without attaching, spawn a new agent in a selected group, or revive a dead session into its own conversation.
- Full-file diff review with syntax highlighting: comments left on lines are sent back to the agent's pane as one review prompt.
- Sessions can be forked to continue conversations in separate named forks, and agents can spawn other agents to send messages and wait for completion.
- Mouse support across the entire manager: scroll, click to focus, drag rows to reorder or move to another group, and open row actions from the […] icon.
- Save named instructions in Settings and reuse them from the New Session form, quick prompt, focused session, spawn, send, or create_session commands.
Use cases
- Managing multiple AI coding agents in a single terminal view without switching between tabs or windows.
- Reviewing diffs generated by agents line by line and sending consolidated feedback back into the agent session.
- Building workflows where agents cooperate: one agent spawns another, sends a message, and waits until the other finishes.
- Keeping a shell alongside any agent session for builds, git operations, and one-off commands.
Getting started
Install using Homebrew on macOS/Linux: bash brew install agent-manager On Windows via Scoop: powershell scoop bucket add yoanwai https://github.com/YoanWai/scoop-bucket scoop install yoanwai/agent-manager For macOS/Linux install script: bash curl -fsSL -o install.sh https://raw.githubusercontent.com/YoanWai/agent-manager/main/install.sh && sh install.sh For Windows install script: powershell irm https://raw.githubusercontent.com/YoanWai/agent-manager/main/install.ps1 -OutFile install.ps1 if ($?) { Get-Content .install.ps1 Unblock-File .install.ps1 powershell -NoProfile -ExecutionPolicy Bypass -File .install.ps1 } See the repository README for more details.
License and pricing
Licensed under Apache-2.0. This is an open source tool with no pricing stated.
Based on the project's public README and repository metadata.