Daintree Assistant
An agent that drives your habitat
Daintree runs a fleet of AI coding agents in parallel, each in its own worktree. The Assistant is the part that drives the fleet for you — it spawns terminals, watches their output, and takes the next action based on what it sees.
you fan the review task out across all five worktrees
▸ Spawned 5 Claude Code terminals
feat/auth · feat/billing · fix/cache · feat/search · fix/ratelimit
▸ Watching for state changes…
you which ones are waiting on me?
▸ Read output across the fleet
⚠ feat/billing — “Overwrite migration 0009? (y/n)”
the other four are still working
you dock each one as it finishes
▸ Listening — I’ll move them to the dock on completion.
The Assistant is a feature built into Daintree, not a separate product and not an extra cost. It runs through the agent CLI you already use, against the subscription you’re already logged into — so there’s nothing extra to pay for. It just takes the controls and does what you’d otherwise be doing by hand.
Under the hood
Not a chat box. A real agent session.
The Assistant has the one thing a bare CLI doesn’t: live orchestration context. It sees every panel, worktree, and agent — and can act on them.
- 01
Runs through the CLI you already have
It runs through whichever agent CLI you already use — Claude Code, Gemini CLI, or Codex — against the subscription you’re logged into. The Assistant is just another sandboxed session on that CLI, not a chat box bolted onto the app.
- 02
Attaches to a local MCP server
It connects to an in-process
daintreeserver that exposes over 300 actions at the permission tier you grant, plus adaintree-docsserver so it can answer how-to questions on the side. - 03
Watches the fleet and acts on it
It subscribes to agent state changes and reads terminal output, so it can act the moment something shifts — answer a prompt, resume a rate-limited agent, dock a finished one — without you walking the panels yourself.
What people ask it
The babysitting, handled
These are the jobs you used to do by hand across a fleet of agent terminals. Now you just ask.
you Fan the review task out across all five worktrees.
Spawns an agent in each worktree and sends them the same prompt, a few at a time, then reports back when they’re all running.
you Find the agents that hit their API limit and resume them once it clears.
Reads each terminal’s output, spots the rate-limit errors, waits out the window, and sends every one of them back to work.
you Which agents are waiting on a question?
Polls the whole fleet in one pass and tells you who’s blocked and on what — the exact prompt each one is stuck on.
you Dock each one as it finishes.
Listens for completions and moves terminals to the dock as they land, so the grid only shows what’s still working.
you Tell all of them to switch to the v2 client and re-run.
Broadcasts the same follow-up to every running terminal at once instead of you typing it into each panel.
you Restart the ones that failed.
Reads the errors, kills the dead terminals, and relaunches them in the same worktree where they left off.
You stay in control
It reaches exactly as far as you let it
The Assistant only goes as far as the permission tier you grant. Set it per session, switch it any time, and it asks before anything destructive.
Workbench
Watches everything, changes nothing. Reads panels, worktrees, and agent state.
Action
Spawns agents, broadcasts prompts, reads terminal state, closes terminals. Nothing destructive.
System
Adds what changes your repo and machine: delete worktrees, commit and push, open issues and PRs.
Let the Assistant run the fleet
Free, open source, no account required. Download and run it locally.