v0.25.2 — You stay signed in across app restarts.

The workspace for your
agents and dev stack

Stop juggling terminals. One workspace where AI agents and dev tools actually work together. Config-driven, with process monitoring built in.

macOS
Windows
Linux
How it works

Three steps, no magic

1

Open a project

Add a project folder. HiveTerm detects your stack automatically.

2

Set up your agents

Add agents and commands. Configure in the UI or write a hive.yml.

3

Let them work

Run hv swarm. Agents coordinate via MCP, notify you when they need you.

Agent coordination

Agents that actually coordinate

8 MCP tools. Claude spawns a test runner. Codex reads its output. The build agent pings you when it's done. They figure it out.

Agent spawns a sub-agent

Claude finds a failing test, spins up a dedicated fixer agent, and keeps working on its own task.

Agents read each other's output

One agent checks your dev server logs, spots an error, writes a fix, and restarts the process. You didn't have to ask.

You get notified, not interrupted

Long build running? The agent sends a native desktop notification when it's done. You keep working on something else.

NEW IN v0.16.0

Talk to your agents

Press a key, say what you want, and the text shows up in the agent's input. Useful when you've got an idea but don't feel like typing two paragraphs.

  • You drop in an OpenAI or Groq key. The audio goes straight there.
  • It auto-stops when you stop talking. You set how long the silence has to be, or disable it.
  • 11 languages, plus auto-detect. Or just tell it to use whatever language the app is in.
  • The transcription lands in the input. Nothing gets sent until you hit enter, so you can clean it up first.
M or click the floating mic in any agent terminal
0:00
Features

What it actually does

One window, all processes

Agents, dev servers, watchers, tests. Split panes, tab switching, all running in real terminals with full PTY support.

One YAML, whole environment

Define your bees in hive.yml. Commit it. Your teammate clones the repo, runs hv swarm, and gets the exact same setup.

Built-in MCP server

Agents talk to each other through a local MCP server. They spawn sub-agents, read each other's output, send you desktop notifications.

Crash recovery and monitoring

Processes auto-restart on crash. CPU and memory usage per process. Output buffering so you can check what a headless agent did while you were away.

VS OTHERS

How it stacks up

Feature HiveTerm Terminal + tmux
AI Agent Workspace
Config-driven setup (hive.yml)
Built-in MCP Server
Sub-agent spawning
Process monitoring & auto-restart
Split panes
Desktop notifications from agents
Cross-platform (Mac/Win/Linux)
Team config sharing
Pricing

Free to start. Pro when you need more.

Free

Enough to run a real project with agents.

$0 /forever
Download Free
  • Up to 2 projects
  • 3 bees per project
  • 6 total bees
  • 1 MCP sub-agent
  • Full terminal & PTY support
  • Config-driven setup

Pro

MOST POPULAR

No limits. Run as many projects, bees, and sub-agents as you want.

$99 /year
Get Pro
  • Unlimited projects
  • Unlimited bees per project
  • Unlimited total bees
  • Unlimited MCP sub-agents
  • Everything in Free
  • 2 machine activations

Try it. It's free.

Download, write a hive.yml, run hv swarm. You'll know in 60 seconds if this is for you.