cmux-browser

Automate end-to-end browser interactions on cmux surfaces using refs and actions.

Updated Mar 21, 2026
One-click install
npx skills add https://github.com/Derek-X-Wang/skills --skill cmux-browser-derek-x-wang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cmux-browser
Source: https://github.com/Derek-X-Wang/skills/tree/main/cmux-browser
Command: npx skills add https://github.com/Derek-X-Wang/skills --skill cmux-browser-derek-x-wang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Enables automated browser interactions inside cmux webviews by coordinating surface contexts, element refs, and action sequences to eliminate manual testing and repetitive tasks.

Core Features & Use Cases

  • Open and target distinct cmux browser surfaces to run isolated automation sessions.
  • Snapshot and manage element refs with a stable workflow to perform clicks, fills, types, and waits.
  • Re-snapshot after navigation or DOM changes to prevent stale references and ensure reliability across tasks, demos, or automated tests.
  • Use ready-made templates and references to build repeatable browser automation pipelines, such as login flows, form automation, or data extraction.

Quick Start

Open a cmux browser surface, take an interactive snapshot to obtain refs, then use those refs to perform actions and wait for state changes.

Frequently Asked Questions about cmux-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate browser interactions in isolated webview surfaces?

Browser automation in isolated webview surfaces is achieved by opening distinct cmux browser surfaces and running separate automation sessions to prevent context conflicts. You target each surface independently to execute action sequences without interference.

How do I prevent stale element references during webview automation?

To prevent stale element references during webview automation, you take an interactive snapshot to obtain refs, perform actions, then re-snapshot after any navigation or DOM changes. This ensures refs remain valid throughout the workflow.

What is the best way to build repeatable browser automation pipelines for login flows?

Building repeatable browser automation pipelines for login flows uses ready-made templates and references to coordinate clicks, fills, and waits. You capture element refs via snapshots and sequence actions across isolated surfaces for reliable execution.

Does cmux browser automation require CLI workflows for end-user tasks?

Cmux browser automation for end-user tasks relies on a cmux CLI workflow to coordinate surfaces, refs, and actions. This CLI integration supports optional references, templates, and surface-based state management for reliable automated execution.

Can I run automated form filling and data extraction across multiple browser surfaces?

Automated form filling and data extraction across multiple browser surfaces is supported by opening distinct cmux surfaces and managing element refs through snapshots. You coordinate isolated sessions to perform fills, clicks, and waits simultaneously.

Why do my browser automation actions fail after page navigation in cmux webviews?

Browser automation actions fail after page navigation in cmux webviews because DOM changes invalidate previously captured element refs. Re-snapshotting after navigation or state changes refreshes refs and restores reliable action execution.