Can I mock API responses with this skill?

Absolutely. The network routing feature allows you to intercept specific URL patterns to abort requests or provide custom mock response bodies for testing edge cases.

Does it support both headless and headed modes?

Yes, while it usually runs in the background, you can use the --headed flag to visually debug and see the actual browser window during execution.

How does Agent Browser identify elements?

It uses a snapshot system that generates unique reference tags (like @e1) for interactive elements based on the accessibility tree, making it easy for the AI to target specific UI components accurately.

Is video recording supported for debugging?

Yes, you can start and stop recordings to capture browser sessions in .webm format, which is ideal for documenting bugs or verifying complex automation flows.

Can I use Agent Browser for authenticated sites?

Yes, it supports saving and loading authentication states via JSON files, allowing you to maintain login sessions and cookies across different tasks.

Agent Browser

Name: Agent Browser
Author: shipshitdev

byshipshitdev

•

Browser Automation

Automates web browser interactions for end-to-end testing, data extraction, and complex web navigation tasks.

Agent Browser is a comprehensive Claude Code skill designed to empower AI agents with full control over web browsers via a specialized CLI. It enables automated navigation, interactive element selection through a unique reference system, form filling, and visual validation via screenshots or video recording. This tool is particularly useful for developers needing to perform QA testing, automate repetitive web-based workflows, or extract structured data from dynamic web applications where standard scraping might fail, providing a robust interface for both semantic and coordinate-based interactions.

Key Features

01Semantic locators for robust element finding by role, text, or label

02Full-page screenshots, PDF generation, and WebM video recording

03Network request interception, request blocking, and API response mocking

04Interactive element referencing for precise UI interaction via unique @refs

0510 GitHub stars

06Support for multi-tab sessions and authenticated state persistence

Use Cases

01Automating multi-step form submissions, checkout flows, and authentication

02Automated End-to-End (E2E) testing and visual regression suites

03Extracting data from dynamic, JavaScript-heavy websites that require interaction

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add shipshitdev/library agent-browser

For use in Claude.ai and ChatGPT

Download Skill