---
title: "Giving Coding Agents Eyes"
description: "Coding agents that produce visual output need a way to look at what they made. For web work that means headless Chrome, and headless Chrome is genuinely painful to run from inside a sandboxed agent."
url: https://ai.statico.io/2026/05/29/giving-coding-agents-eyes/
canonical: https://ai.statico.io/2026/05/29/giving-coding-agents-eyes/
date: 2026-05-29T16:11:05.000Z
last_updated: 2026-05-29T16:11:05.000Z
doc_version: 2026-05-29
tags: ["agents", "chrome", "claude-code", "coding-agents", "headless-browser", "macos", "playwright", "sandboxing", "tooling", "web"]
ai_disclosure: ai-assisted
---

# Giving Coding Agents Eyes

Coding agents that produce visual output need a way to look at what they made. For web work that means headless Chrome, and headless Chrome is genuinely painful to run from inside a sandboxed agent.

Chromium and Firefox both rely on Mach-O quirks, macOS entitlements, and Crashpad behavior that don't survive most sandboxes. I run my agents inside [nono.sh](https://nono.sh) profiles per project, and Chrome under that setup is a non-starter.

## The workaround

Playwright runs fine outside the sandbox. So it lives on a high port and Claude is told, in its instructions, to always talk to the Playwright MCP server there:

```
$ npx @playwright/mcp@latest --headless --isolated --browser chrome --port 8931
```

The sandbox just needs to reach `localhost:8931` and the visual-review loop works. Claude renders the local service, takes a screenshot, looks at it, iterates.

That mostly works. What it does not solve: stale processes, hanging Chrome instances, zombies. Every so often Chrome spins out and eats *all 64 GB of RAM* on my M4 MacBook Pro before I notice.

## Lighter options

There has to be something simpler than babysitting a browser. Two things caught my eye recently.

[Webwright](https://www.microsoft.com/en-us/research/articles/webwright-a-terminal-is-all-you-need-for-web-agents/) from Microsoft Research gives the model a terminal and a workspace, and lets it write Playwright code that launches, inspects, and discards browser sessions. The output is a reusable script, not a chat transcript. It scores 60.1% on Odysseys against base GPT-5.4's 33.5%, which is a real jump.

[obra/superpowers-chrome](https://github.com/obra/superpowers-chrome) goes the other direction: a Claude Code plugin that drives Chrome directly via the DevTools Protocol, zero dependencies, no Playwright in the middle.

## When you actually need real Chrome

Advanced bot fingerprinting is the case for keeping a full browser around. If the task is logging into a hostile site or completing a real-world flow, real Chrome with a real profile is the only thing that works.

But most of my use is smaller: render a local dev server, screenshot it, ask Claude if the layout looks right. For that, a 64 GB RAM-eating Chromium feels like the wrong shape of tool. I suspect this gets cleanly solved within a year, probably by something CDP-direct and disposable rather than a long-lived browser process I have to nanny.

## Sitemap

- [All posts](https://ai.statico.io/)
- [Markdown sitemap](https://ai.statico.io/sitemap.md)
- [llms.txt index](https://ai.statico.io/llms.txt)
- [Archive](https://ai.statico.io/archive)