# thaytool — features

[home](https://thaytool.com/) › features

six things thaytool does, in plain terms. no metaphors, no orchestra analogies. if you run CLI agents and know what a git worktree is, this will read like documentation — on purpose.

## persistent terminals

**terminals that survive.** most apps kill their terminals when you quit. thaytool runs a detached PTY daemon in the background. close the app, reopen it — your agents are still running, scrollback and all, attached back exactly where they left off. not a tmux wrapper. the actual process, still alive, waiting for you.

## a git worktree per task

**one task. one worktree. no collisions.** each workspace is a real `git worktree` — a separate working tree checked out from the same repo, on its own branch from the start. agents can't step on each other because they're not in the same directory. when one's done, you review the diff and merge. main stays clean until you say so.

## agents in parallel

**run the whole team at once.** claude code on the auth refactor. codex on the test suite. claude code again on the api migration. three agents, three worktrees, one window. status is inferred from the actual foreground process — not a guess, not a log line.

## real system data

**real data. not a dashboard. not a simulation.** cpu from `mach`. listening ports from `lsof`. git status from the actual working tree. thaytool reads the OS directly — no agent to poll, no server to query. dev server on :3000? you see it. an agent pegging a core? you see it. uncommitted changes in a worktree? you see it.

## native macOS, single binary

**SwiftUI. macOS 26 Liquid Glass. one binary.** no electron. no node runtime bundled inside a shell script. no update daemon running in the background you forgot about. it ships as one binary, behaves like a Mac app because it is one. uninstall it and it's gone.

## 100% local

**your work stays on your machine.** no cloud backend. no auth server. no account to create. no api key to revoke if the company pivots. your repos, terminals, and agent sessions never leave the machine — not because of a privacy policy, but because there is nowhere for them to go. single-user. fully local. deliberately scoped.

## how it works

1. **add a repo** — point thaytool at a git repo on your disk. it reads the current state — branches, worktrees, uncommitted changes. that's the whole setup.
2. **create a workspace** — pick a task, name a branch. thaytool checks out a new git worktree for it, sandboxed so it can never touch your main checkout, and opens a persistent terminal inside it.
3. **run your agent** — drop into the terminal and start claude code or codex. switch to another workspace and do it again. thaytool tracks diffs, ports, and metrics while you work.
4. **review, or walk away** — review the diff and merge when you're ready. or close the laptop and come back tomorrow — the terminals will still be running.

## questions

### does this work with claude code and codex?

yes. thaytool opens a standard terminal in each worktree. any CLI agent that runs in a terminal works — claude code, codex, aider, whatever you use.

### what happens if i close the app?

your terminals keep running. a detached PTY daemon owns the sessions and outlives the GUI. reopen the app and sessions reattach with scrollback replay.

### is there a cloud / hosted version?

no, and there won't be. thaytool is deliberately scoped to the local machine. that's the point.

### does it run on linux or windows?

no. it's a native macOS app and uses macOS-only APIs (forkpty, libproc, mach, Liquid Glass).

See [install](https://thaytool.com/install).

---

_Last updated: 2026-05-31 · https://thaytool.com_