dogfood

Explores web applications via browser automation to find bugs and produce evidence-backed issue reports.

1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/Alwahdi/mikrotik-whisperer --skill dogfood-alwahdi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dogfood
Source: https://github.com/Alwahdi/mikrotik-whisperer/tree/main/.agents/skills/dogfood
Command: npx skills add https://github.com/Alwahdi/mikrotik-whisperer --skill dogfood-alwahdi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agent-browser, and includes references (resource) components.

What problem does it solve? Manually testing a web application for bugs, UX problems, and console errors is time-consuming, and findings are often reported without enough evidence for engineers to reproduce them. This Skill systematically explores a web app like a real user and documents every issue with screenshots, repro videos, and step-by-step instructions. ## Core Features & Use Cases - Systematic Exploration: Walks through navigation, forms, interactive elements, edge cases, and end-to-end workflows using the agent-browser CLI. - Repro-First Documentation: Captures annotated screenshots and repro videos for interactive issues, and single annotated screenshots for static issues like typos or misalignment. - Structured Reporting: Produces a severity-ranked report (critical/high/medium/low) using a consistent template with an issue taxonomy covering visual, functional, UX, content, performance, console, and accessibility categories. - Use Case: Point it at a staging URL before a release, and receive a report of 5-10 well-documented issues that can be handed directly to engineering and design teams. ## Quick Start Dogfood the app at http://localhost:3000 and give me a report of all the issues you find.

Frequently Asked Questions about dogfood

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

FAQPage Schema
How do I dogfood or exploratory test a web application?

Provide the target URL and the Skill launches an agent-browser session, navigates each section, clicks interactive elements, fills forms, and checks the console. Every issue found is documented immediately with screenshots or video and written into a structured report.

What kinds of issues does exploratory web testing find?

It covers seven categories: visual/layout defects, functional bugs like broken links and failed forms, UX problems like missing feedback, content errors like typos, performance issues, console errors and failed requests, and accessibility gaps like missing labels.

Can I test a web app that requires login?

Yes. Provide credentials and the Skill signs in through the browser, handles OTP or email codes by asking you, then saves the auth state to a JSON file for reuse during the session.

What evidence is included for each bug found?

Interactive issues get a repro video plus step-by-step screenshots mapped to numbered repro steps. Static issues like typos or misalignment get a single annotated screenshot. Each issue also records severity, category, and page URL.

Why use agent-browser directly instead of npx agent-browser?

The direct agent-browser binary uses a fast Rust client, while npx routes through Node.js and is significantly slower. The Skill instructs using the direct binary for all browser commands.