avad-browse

Drive a persistent headless browser for QA testing and site dogfooding.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/agwacom/avadbot --skill avad-browse
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: avad-browse
Source: https://github.com/agwacom/avadbot/tree/main/skills/avad-browse
Command: npx skills add https://github.com/agwacom/avadbot --skill avad-browse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

avad-browse provides a persistent headless browser that Claude Code can drive for QA testing and site dogfooding, preserving state across commands for fast, repeatable checks.

Core Features & Use Cases

  • Persistent browser daemon with cookies, tabs, and login sessions retained between commands, enabling rapid iterative testing.
  • Rich command surface for navigation, interaction, inspection, visualization, and state capture (diffs, annotated screenshots, and performance data).
  • Use Case: QA teams can automate a full user flow through a web app and generate evidence (screenshots, diffs, console/network logs) for bug reports.

Quick Start

Run the avad-browse CLI to launch a persistent headless browser and begin issuing commands like goto, click, and snapshot to verify behavior.

Frequently Asked Questions about avad-browse

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

FAQPage Schema
How do I automate QA testing with a persistent headless browser?

Automate QA testing by launching a persistent headless browser daemon that retains cookies, tabs, and login sessions between commands, enabling rapid iterative testing and reliable end-to-end QA runs across environments.

What is snapshot diff testing and how does it help with site dogfooding?

Snapshot diff testing captures and compares visual state changes during site dogfooding, generating annotated screenshots and performance data to verify behavior and collect evidence for bug reports.

Can I retain login sessions and cookies between browser automation commands?

Yes, the persistent browser daemon retains cookies, tabs, and login sessions between commands, preserving state across interactions for fast, repeatable QA checks without needing to re-authenticate.

How do I capture console and network logs for bug reports during automated testing?

Capture console and network logs by driving the headless browser through a full user flow, generating evidence like screenshots, diffs, and performance data for comprehensive bug reports.

Does avad-browse work with Playwright for end-to-end QA workflows?

avad-browse provides a persistent headless browser applicable to automated QA workflows, manual testing automation, and evidence collection, satisfying safety checks, health monitoring, and state persistence for reliable runs.

What are the limitations of using a headless browser for manual testing automation?

Headless browser testing lacks a visible UI for visual verification, requiring snapshot diffs and annotated screenshots to inspect behavior, and may not fully replicate user interactions with graphical rendering.