Handler is a Mac app for Codex and OpenCode that adds a review layer while the agent is generating code.
Every edit comes with a short explanation: what changed, why it changed, and what it touches. If one edit needs more context, you can open a side chat on just that change and ask follow-ups without derailing the main agent run.
The goal is to build a mental model as the code is generated, instead of waiting until the end and reviewing one giant diff cold.
It’s the same idea that makes stacked PRs easier to review: smaller, understandable units. Handler applies that at AI generation time, before the change turns into one giant diff.
This year I mostly stopped writing code by hand. I drive agents now, and I’m faster for it.
But a few months in I noticed a problem. The agent would come back with a 400-line change, tests green, diff looked reasonable, and after the first few chunks I’d pretty much lose the mental model. At that point I wasn’t really reviewing anymore. I was skimming the shape of the change and moving on.
I know that’s not the ideal way to drive an agent. But I also don’t want to slow the whole thing down and manually inspect every line like it’s 2022. The whole point of using these tools is speed.
So the bottleneck stopped being writing code. It became understanding the code fast enough to stay in control of my own project.
Handler is what I built to fix that for myself:
- Every edit the agent proposes comes with a plain explanation of what that specific change does and why. - The explanation is attached to the edit, so you build context incrementally instead of reading one huge summary at the end. - You can pull any single edit into a side chat and ask “why this approach?”, “what else does this affect?”, or “redo this part (soon)” without blowing up the main agent context. - You go edit by edit and accept or reject, instead of staring at one wall of diff and hoping. - It runs on your existing Codex and OpenCode setups, with worktrees so parallel runs don’t collide. - It also has a built-in terminal that feels closer to a Chrome DevTools console than a plain terminal: command output is easier to inspect, the agent can read the terminal state, and there’s a native JSON viewer/builder for working with structured data without copy-pasting into another tool.
What it’s not: a review bot like CodeRabbit or Greptile. I use those, and they catch real bugs. But they review the code after the fact.
Handler points earlier in the workflow. It helps you review while the agent is producing the change, so you can keep the mental model instead of trying to reconstruct it later from a finished PR.
I’ve been using it daily and can’t go back to merging agent changes blind. If you’ve felt the same problem with large AI-generated diffs, I’d love for you to try it and tell me where it falls over.