Aibex
Writing < Reading, Thinking, Deciding
Aibex is a desktop app for reading your agents' work.
In development · macOS desktop app · Free beta in preparation · Paid full release to follow
The problems it sets out to solve
Developing with AI is stressful for reasons it never used to be.
- Every agent turn comes with a wait.
- Several projects run at the same time.
- You have to read through large amounts of text and code to review it.
- Every switch between tasks carries a steep context-switching cost.
- A task finishes, and you do not notice.
What already works
Select diff lines, then ask
Select diff lines and press "Attach to question" — your question carries those lines with it. No more re-explaining "around line 42 of that function". You can also leave a comment on the line yourself.
Pick the working folder four ways
- Starred bookmarks
- An automatic list of recent folders
- A browser for folders whose names you do not know
- Plain path entry for the ones you do
One look at the tab tells you
- The project name
- The AI's state (Idle, Running, Your turn or Crashed)
- Memory use
"Your turn", presented in order
When an agent finishes replying, its tab joins a queue as awaiting you. Send and go next sends your reply and jumps to whichever tab has been waiting longest; with nothing to add, it simply moves you on. Which tab needs you is not something you have to remember.
Tells you when it is done
An elapsed-time strip stays on screen while the agent runs, and a sound marks the finish. It cuts down the trips back to the window just to check whether it is done yet.
Why it changed, on record
The agent posts its own reasoning as an inline comment on the affected diff line, so the motive for a change is not buried in the transcript. It can be switched off.
No more switching between terminal and chat
When the agent says "please run this command in a terminal", select it and choose "Run in terminal" from the right-click menu. That is all. The command runs under your own permissions, and its output and exit code stay in the conversation and reach the agent with your next message.
Price
A free beta comes first; after it, Aibex moves to a paid full release.
| Beta | The beta is free. It is still being prepared. Each beta stops working 90 days after it was built; a newer one arrives by automatic update before then. |
|---|---|
| Full release | The full release will be paid. The price and the length of the trial will be announced on this page once they are set. Beta users will need to buy it too once the full release is out. |
Where it stands
| Availability | A free beta is on its way via GitHub Releases. |
|---|---|
| Platform | It runs only on Macs with Apple silicon. Intel Macs are not supported. The desktop app is being prepared. |
| Construction |
The app starts a small server on your own machine and shows its interface in a window.
It listens on 127.0.0.1 only, so nothing outside that machine — including
other devices on your network — can reach it. It uses no frameworks.
|
| AI |
It uses Claude through the Claude Agent SDK. Using it requires a Claude subscription.
Settings are read from both ~/.claude/settings.json (user-wide) and the
working folder's .claude/settings.json (project).
|
| Price | The beta is free. The full release will be paid; the price is not set yet. |
What you need to get started
Aibex does not bundle the agent itself. It calls the Claude already installed on your machine. If you are using Claude Code today, items 1 and 2 are already in place. Check the version of item 3, Node.js, and whether you have item 4, pnpm.
| 1. A Claude subscription |
Your own subscription is used as-is. There is no API key to register with Aibex,
and you are not billed twice for the AI. Sign in with claude login in a
terminal before you start.
|
|---|---|
2. Claude Code (the claude command) |
Running the agent is left to this command. On launch the app looks in
~/.local/bin and the Homebrew locations in turn, and shows a page
explaining how to install it if nothing is found.
|
| 3. Node.js 22.15 or newer |
Used to run the small local server the app starts on your machine. Like
claude, it is located automatically at launch.
In the 22 line that means 22.15 or later, in the 23 line 23.5 or later, and any 24 or later.
23.0 through 23.4 do not work, even though they are newer than 22.15 —
the feature it needs is missing from that range.
|
| 4. pnpm |
Used to perform the first-launch download below. It can be enabled through Corepack,
which ships with Node.js, using corepack enable pnpm — or installed with
brew install pnpm if you use Homebrew.
|
| A first-launch download |
The first time you open it, your machine fetches
@anthropic-ai/claude-agent-sdk — the npm package Anthropic publishes, which
the app uses to call Claude — directly from the official npm registry.
We are not redistributing it.
Once installed it takes up around 220 MB on disk. You are shown exactly this before the
fetch begins, and later launches do not fetch again (except to update the package).
|
| Network | The agent's reasoning runs on Claude, so a connection is required whenever you use it. Reading and writing files and running commands all stay on your own machine, but the agent cannot run offline. |
Feedback and bug reports
Issues are welcome. Bug reports, questions, "it does not run on my machine" — every one of them is read, and so are notes on the awkwardness you would like fixed.
Questions
When can I use it?
A built app is being prepared as a free beta; once it is on GitHub Releases, the "Where it stands" table above links to it. It is built while being used for its author's own daily development, so its shape keeps changing as the awkwardness gets easier to name.
Is it a web service that runs in a browser?
No. It only uses browser technology for its interface; it is an application that runs on your machine. File access and command execution happen entirely locally (the agent's reasoning itself uses Claude).
Does my existing Claude Code configuration carry over?
Yes. The sandbox boundary and deny rules come straight from both
~/.claude/settings.json (user-wide) and the working folder's own
.claude/settings.json (project). settings.local.json is
deliberately not read. Your user-defined slash commands
(.claude/commands/) are picked up as well.
Does putting this app in the middle weaken the sandbox?
No. How far a command may reach is left to the OS sandbox, and that reach is set by
~/.claude/settings.json and the working folder's
.claude/settings.json — the app never parses command strings itself. On top
of that, the app refuses any file-writing operation by the agent the moment its path
points outside the working folder, and that check is applied twice so it still holds
when the permission mode is relaxed.
Will other AI models be supported?
Today it assumes the Claude Agent SDK. A quick model switch is logged as a known gap, but work on it has not started.
How does it relate to the Draft apps?
It is a separate product. It does not integrate with DraftFinder, DraftCamera or DraftWriter.
If you want to follow along
When the beta lands on GitHub Releases, this page will say so.