Claude Code gains computer use preview

// 57d agoPRODUCT UPDATE

Claude Code gains computer use preview

Anthropic has added computer use to Claude Code as a research preview for Pro and Max plans on macOS. The CLI agent can open apps, click through the UI, and test what it built, pulling verification into the same build-and-test loop.

// ANALYSIS

This is the kind of update that turns Claude Code from a coding assistant into a QA co-pilot. Anthropic is keeping the blast radius small: Mac-only, Pro/Max-only, and wrapped in prompt approvals.

–Native-app and simulator testing are the killer use cases here: compile, launch, click, screenshot, iterate.
–It ships as a built-in `computer-use` MCP server, and Claude falls back to it only after MCP, Bash, and Claude in Chrome.
–Anthropic says the feature can be slow and mistake-prone, which is a good reminder that this is supervised automation, not unattended autonomy.
–The macOS-only gate means Windows and Linux users are still watching from the sidelines.

// TAGS

claude-codeai-codingagentcomputer-useclitestingautomationmcp

DISCOVERED

57d ago

2026-03-30

PUBLISHED

58d ago

2026-03-30

RELEVANCE

9/ 10

AUTHOR

policyweb

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE4h ago

Cursor adds dedicated subagents for skills

Cursor now allows developers to execute tool-heavy or research-intensive agent skills within dedicated subagents. This architectural shift isolates noisy background tasks, keeping the main chat context clean and focused.

UPDATE4h ago

YouTube moves AI labels to video player

YouTube is moving its AI content disclosures from video descriptions to more prominent placements beneath the player and on Shorts overlays. Starting in May, the platform will use internal signals to automatically label photorealistic AI content that creators fail to disclose.

OPEN SOURCE7h ago

Taste Skill kills AI "frontend slop"

Taste-Skill is an open-source framework that provides portable "agent skills" to enforce high-end design principles in AI-generated code. By injecting specific design directives and "anti-slop" rules, it enables LLMs to produce editorial-grade UIs that bypass generic, boilerplate-heavy AI templates.