Treebar: Manage Git Worktrees from Your Mac Menu Bar and Keep Every AI Coding Agent in Sight

When AI Coding Meets the Multi-Worktree Problem

As AI coding assistants like Codex and Claude Code become more widely adopted, developers are increasingly turning to "multi-worktree (Git Worktree)" workflows — letting AI Agents work across multiple branches in parallel. But that efficiency gain comes with a new pain point: when your Codex is running across five worktrees simultaneously, it's hard to know at a glance where it is, what it's changed, or how many Agents are actually running.

That's exactly the problem Treebar aims to solve. It's a macOS menu bar app that surfaces the status of all active worktrees in your MacBook's notch area, giving you a full picture without switching terminals or your IDE.

Git Worktree is a native Git feature introduced in Git 2.5. It lets developers check out multiple branches from the same local repository into separate filesystem directories simultaneously — each directory being an independent "worktree." In traditional workflows, switching branches with git checkout requires stashing uncommitted changes and breaks your flow. With multiple worktrees, each branch lives in its own directory, so you can work across several terminals or IDE windows in parallel without interference. In AI coding scenarios, this means different Agent instances can handle different feature branches at the same time — one Agent works on feature A, another fixes bug B. The efficiency gains are real, but for the human overseeing all of this, information scattered across multiple directories and terminal sessions multiplies the cognitive load.

Treebar's Core Feature: One View for All Git Worktrees

Treebar's core value is a unified view. It monitors Git and Codex activity in real time, aggregating the state of every worktree into one place. Developers can see three key dimensions at once:

• Where: which worktree the AI is currently operating in
• What: which files are being changed
• How many: how many Agents are running in parallel

The clever design choice here is placing this information in the MacBook notch — always visible, but not occupying any main screen real estate. Compared to cycling through terminal windows or tabs, this persistent, lightweight display fits naturally into a developer's peripheral awareness.

A Workflow Built Around AI Coding with Codex

Notably, Treebar isn't a general-purpose Git monitoring tool. It's built specifically around AI coding workflows, with dedicated support for tracking Codex activity — understanding where an AI Agent is working and what it's doing. This reflects a sharp insight from the author: a new development paradigm is emerging where human developers shift from "writing code themselves" to "supervising multiple AI Agents working in parallel." And effective supervision requires observability.

Observability is a foundational concept in distributed systems — the ability to infer a system's internal state from its external outputs (logs, metrics, traces). In the context of AI Agent coding, the concept maps naturally: when multiple Agents are autonomously modifying code in the background, developers need to answer "where is the Agent right now, what has it done, and does the output match expectations?" These are the three pillars of observability — tracing (location), metrics (state snapshots), and logging (change history) — applied to a new domain. Running a multi-Agent development setup without observability is like operating a microservices cluster with no monitoring dashboard. The risk isn't that the Agents aren't smart enough; it's that the humans have lost meaningful oversight.

A Tool Born from a Real Personal Pain Point

According to the author, Cory Parry, Treebar was actually built a few months ago for a simple reason: he kept losing track of where Codex was working. Products born from genuine personal frustration like this tend to feel grounded — they're solving a friction point that comes up over and over in daily development, not a hypothetical use case.

This "scratch your own itch" style of indie development is a common source of quality products on Product Hunt. Treebar launched with 68 upvotes and ranked 18th for the day — not a breakout hit, but a solid showing for a focused menu bar utility targeting a niche workflow.

Core Code Going Open Source

The author also shared an important upcoming plan: the core of Treebar will be open-sourced soon. That's a positive signal for the developer community.

Open-sourcing the core does a few things: developers can audit a tool that reads Git and code change data, which eases privacy and security concerns; the community can extend support to more AI coding tools (like Claude Code, Cursor, etc.) beyond just Codex; and it lowers the cost for developers with similar needs who would otherwise build something from scratch.

A Signal of Where AI-Assisted Development Is Heading

Treebar is a small menu bar utility, but it reflects a deeper shift in the AI coding era: developers are moving from "writing" to "orchestrating and supervising."

When one person can drive multiple AI Agents developing across different branches in parallel, the new core challenge becomes managing that parallelism and maintaining meaningful control over AI behavior. Tools like Treebar are essentially filling in the observability gap in human-AI collaborative coding. More tools focused on AI Agent monitoring, orchestration, and auditing are likely to follow, forming an entirely new tool category.

For macOS users who rely heavily on Codex and are comfortable with multi-worktree workflows, Treebar is a lightweight productivity tool worth keeping an eye on — especially once its core goes open source, which opens up a lot of possibilities.