← Back to Discover

Cate

Open-source desktop workspace that places terminals, editors and browsers on a zoomable canvas, with status tracking for coding agents such as Claude Code and Codex running in its terminals.

Code editorsCoding agents

What it does

Cate is an open-source desktop application, released under the MIT licence, that arranges terminals, Monaco code editors, browsers, document viewers and nested canvases as panels in a spatial workspace. Panels can float on the canvas, dock into tabs and splits, or be detached into their own window, and the layout is remembered per project. It is built around running coding agents in parallel. Agent command-line tools such as Claude Code or Codex run inside Cate terminals, and supported agents report when a turn starts, ends or needs permission, so each terminal shows whether its agent is running, waiting or finished and sends a notification when it needs an answer. Agent sessions come back with their scrollback after a restart. For parallel branches, Cate can create a git worktree and branch from a local branch, a remote branch or an open pull request, giving each its own coloured area on the canvas. It also includes multi-repo source control, side-by-side diffs, ripgrep search and a command palette. A cate command-line tool lets agents drive a browser panel, read another terminal and open files. Workspaces can run on the local machine or on another one, and a second device can join a live workspace over an end-to-end encrypted connection using a QR code or pairing code. Prebuilt downloads are listed for macOS, Windows and Linux.

Capabilities

  • Terminals, editors, browsers and viewers as floating or docked panels
  • Agent-aware terminals showing running, waiting or finished status with notifications
  • Git worktree creation for parallel agent branches
  • Multi-repo source control, side-by-side diffs and ripgrep search
  • A cate command-line tool that agents can call to drive panels
  • Local or remote workspaces with encrypted device pairing

When to use it

  • Running several coding agents on separate branches at once
  • Keeping terminals, editors and a browser together in one spatial layout per project
  • Seeing at a glance which agent is waiting for your input