The agent workspace in your shell.
fluidstate is the fast execution surface. It keeps agentic coding close to the repo, close to your keyboard, and clear enough to supervise multiple agents without opening a stack of terminal tabs.
api agent
$ codex run task-1
reading repo graph...
writing patch
ui agent
$ codex run task-2
reading repo graph...
writing patch
tests
$ codex run task-3
reading repo graph...
writing patch
shell
$ codex run task-4
reading repo graph...
writing patch
fluidstate
What it owns
For builders who already live in the terminal and want agentic work to feel native instead of bolted onto a chat window.
Parallel agent panes with real PTY output
Claude Code, Codex, and shell sessions side by side
Keyboard-first diffs, approvals, and command flow
Workflow
How fluidstate works
01
Install
One binary. Start in any repo.
Run the installer, open a project, and launch agent panes without adding a Node or Electron runtime to the workflow.
02
Fan out
Split work across agents.
Give API, UI, tests, or review work to separate panes while keeping every session visible and interruptible.
03
Land
Review the exact changes.
Stay in the shell for diffs, permission checks, command history, and the final judgment calls before code lands.
How it connects
Part of one loop.
fluidstate is the execution engine. Tree adds architectural visibility to the same workflow, and v2 turns higher-level intent into the runs you dispatch.