dev-browser

Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include "go to [url]", "click on", "fill out the form", "take a screenshot", "scrape", "automate", "test the website", "log into", or any browser interaction request.

3,188stars202forksUpdated 1/23/2026

#chrome #web-scraping #testing #browser-automation #javascript #automation

Security Assessment

Safe(100/100)

Security Score100/100

About dev-browser

The dev-browser skill provides browser automation with persistent page state, allowing users to interact with websites, fill out forms, take screenshots, scrape data, test web applications, and automate browser workflows. This skill is designed to automate repetitive tasks while maintaining page state across script executions, enabling users to build scripts that evolve incrementally. It serves as an essential tool for anyone needing to automate browser-related workflows, whether for testing, scraping, or automation of web-based tasks.

Key features of the dev-browser skill include the ability to use two modes of operation: Standalone Mode (which launches a new Chromium browser) and Extension Mode (which connects to an existing Chrome browser). It supports a variety of browser interactions such as navigation, form-filling, and screenshot-taking. The skill's flexibility allows users to automate complex workflows by breaking them down into smaller scripts that can be tested and refined over time. Additionally, it offers the ability to create named pages, log state information, and work with a dynamic client-server model to handle automation requests.

Target users of the dev-browser skill include web developers, QA engineers, data scrapers, and testers who need a reliable, repeatable method of automating browser tasks. It is particularly useful for those who need to work with dynamic web pages or complex browser sessions that require persistence across multiple script executions.

FAQ

How do I choose between Standalone Mode and Extension Mode?

Standalone Mode is ideal for fresh automation sessions with a new Chromium browser, while Extension Mode is necessary when interacting with an existing browser session, especially if the user is already logged into websites.

Can I use the dev-browser skill for web scraping?

Yes, the dev-browser skill supports web scraping, allowing users to extract data from websites during automation tasks.

What should I do if the extension hasn't connected?

If the extension hasn't connected, ask the user to launch and activate it. You can download the extension from the provided link (https://github.com/SawyerHood/dev-browser/releases).

What are the limitations of running TypeScript in the browser context?

When using `page.evaluate()`, the code runs in the browser environment, which does not support TypeScript syntax. Use plain JavaScript instead.

Can I automate tasks on websites with dynamic content?

Yes, the dev-browser skill is designed to automate tasks even on pages with dynamic content, leveraging tools like `getAISnapshot()` to interact with elements when layouts are unknown.