Skip to content
hiveDOCS
GITHUB ↗v1.0.0-rc.1
Ask Beenot connected

Bee will answer questions from these docs once it has a backend that can answer honestly. It isn't connected yet, so nothing you type here goes anywhere.

integration point
ASK_BEE_ENDPOINT = unset
MEANWHILE
Hand the docs to your own agent

Help me understand and set up Hive. Read https://hive.admir-saheta.com/agent-guide.md first, then walk me through it step by step.

How Hive Works

A practical overview of the Hive daemon, terminal client, environments, and persistent agent processes.

Hive gives command-line agents a persistent place to work. You keep using the coding tools you already know; Hive organizes their terminals, shows what they are doing, and keeps them running when you detach.

The moving parts

text
You → hive CLI / terminal UI → daemon → shim → agent process
                                    └── environments, tabs, panes, snapshots
  • CLI and terminal UI: The hive command opens the multiplexer or runs a management command. Both are clients of the daemon.
  • Daemon: Starts on demand and coordinates Hive’s environments, processes, and saved workspace state.
  • Environment: A named working directory, usually a repository or worktree. Agents in an environment share its project context and environment variables.
  • Tab and pane: Tabs organize layouts; panes display terminal processes in those layouts.
  • Shim: A small supervisor attached to each agent process. It keeps the agent’s terminal and output alive across daemon restarts.

See Core Concepts for the system model and Sessions for how the pieces are grouped.

Start with a project

Open Hive from a repository directory. Create a named environment for it, then start an agent:

sh
cd ~/src/app
hive env create app --cwd .
hive agent start --kind claude --env app

Hive opens the agent in a new tab. Its terminal remains available in the multiplexer, and the sidebar shows its detected state. Start more agents in the same environment or give concurrent changes separate worktrees:

sh
hive agent start --kind codex --env app --worktree fix/login

Keep work moving

Detach with Ctrl+B, then D. The UI closes while the agent continues. Run hive later to return, or use the CLI to inspect output and send work:

sh
hive agent list
hive agent prompt <agent> "Add a regression test for the login redirect" --wait --read 40

Hive can restore a saved workspace after a reboot and resume supported agents’ conversations when their integration hooks are installed. See Persistence.

Choose how you work

You can manage agents interactively in the multiplexer, script the CLI with JSON or templates, or let another agent call Hive through MCP. For work on another machine, Hive can drive a remote daemon over SSH while keeping the interface on your local machine.