Dogfood

This is a systematic approach to exploratory QA testing of web apps using Claude's browser tools. It walks through five phases: planning a test scope, exploring pages and interactions, collecting evidence with screenshots and console logs, categorizing bugs by severity, and generating a structured markdown report. The workflow is thorough about checking JavaScript console errors after every navigation and interaction, which catches silent failures most testers miss. It includes an issue taxonomy and report template. If you need to QA a web app and want a methodical checklist rather than ad-hoc clicking around, this gives you that structure. The emphasis on evidence collection and reproducible steps makes the output actually useful for developers fixing the bugs.

skills.sh installs: 6.4K
GitHub stars: 262.1K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add nousresearch/hermes-agent --skill dogfood --agent claude-code

Installs into .claude/skills of the current project.

Files

Dogfood: Systematic Web Application QA Testing

Overview

This skill guides you through systematic exploratory QA testing of web applications with browser_exec. You will navigate the application, interact with elements, capture evidence of issues, and produce a structured bug report.

Prerequisites

The browser toolset must be enabled (browser_exec). browser_exec runs Python with pre-imported helpers (new_tab, goto_url, wait_for_load, page_info, js, fill_input, click_at_xy, capture_screenshot, cdp); browser state persists across calls, Python variables do not.
A target URL and testing scope from the user

Inputs

The user provides:

Target URL — the entry point for testing
Scope — what areas/features to focus on (or "full site" for comprehensive testing)
Output directory (optional) — where to save screenshots and the report (default: ./dogfood-output)