agent-browser

Automate remote browser interactions via Browserless or CDP endpoints.

35|5|Updated May 7, 2026

Search Tags:#electron #cdp #ai-agents #browser-automation #browserless #qa-testing #remote-backend

Ready to run

npx skills add https://github.com/yokingma/weclaws --skill agent-browser-yokingma

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/yokingma/weclaws/tree/main/resources/skills/managed/agent-browser
Command: npx skills add https://github.com/yokingma/weclaws --skill agent-browser-yokingma

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Browser automation for AI agents requires reliable remote execution, isolation, and centralized control; this skill enables controlling a browser backend from the AI agent platform.

Core Features & Use Cases

Remote-only browser automation with Browserless and explicit CDP endpoints.
Session persistence, authentication vault, and video recording for auditability.
Specialized workflows for Electron apps, Slack, QA, exploratory testing, and cloud providers.

Quick Start

Start by connecting to the remote browser backend with agent-browser -p browserless, then open a URL.

Frequently Asked Questions about agent-browser

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

FAQPage Schema

How do I automate remote browser interactions for AI agents?▼

Remote browser automation for AI agents is achieved by controlling a browser backend via Browserless or explicit CDP endpoints. This enables website navigation, form filling, and screenshots while enforcing remote-only execution with session persistence.

Can I use CDP endpoints to run QA testing on Electron apps?▼

Yes, QA testing on Electron apps is supported through explicit CDP endpoints. The skill controls the remote browser backend to execute exploratory workflows and automated tests while maintaining session persistence and authentication vaults.

Does browser automation work with Browserless for remote-only execution?▼

Browserless is fully supported for remote-only browser execution. Connecting to a Browserless backend allows AI agents to perform cloud provider workflows, Slack interactions, and form filling with video recording for auditability.

How do I connect to a remote browser backend for exploratory testing?▼

Connect to the remote browser backend using the browserless protocol, then open a target URL. This initiates exploratory testing workflows with authentication vaults for secure login and video recording to capture the testing session.

What are the limitations of remote-only browser automation for AI agents?▼

The primary limitation of remote-only browser automation is the strict requirement for a remote backend via Browserless or CDP endpoints. Local browser execution is not supported, requiring consistent network connectivity and external session management.