dogfood

Tests web applications through browser-driven exploratory QA and generates structured bug reports.

6|1|Updated May 11, 2026
One-click install
npx skills add https://github.com/yakeworld/Synthos --skill dogfood-yakeworld
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dogfood
Source: https://github.com/yakeworld/Synthos/tree/main/skills/private/extended/external-automation/dogfood
Command: npx skills add https://github.com/yakeworld/Synthos --skill dogfood-yakeworld

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manual exploratory testing of web applications is slow and inconsistent, and silent JavaScript errors or below-the-fold rendering issues often go unnoticed. This Skill systematizes browser-based QA so every navigation, interaction, and defect is captured with reproducible evidence. ## Core Features & Use Cases - Five-Phase QA Workflow: Plan a sitemap, explore pages with browser tools, collect screenshot evidence, categorize and de-duplicate issues, then generate a structured report. - Silent Error Detection: Checks the browser console after every navigation and key interaction to surface high-value JavaScript errors. - Evidence-Backed Reporting: Every defect is documented with reproduction steps, severity/category classification, and annotated screenshots referenced inline via MEDIA paths. - Use Case: Point the Skill at a local staging app (e.g., a registration flow) and receive a report covering form validation bugs, XSS console errors, duplicate submissions, and layout issues, sorted by severity. ## Quick Start Run a dogfood QA test on http://localhost:3000 covering the signup and login flows, and save the bug report with screenshots to ./dogfood-output.

Frequently Asked Questions about dogfood

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

FAQPage Schema
How do I run automated exploratory QA testing on a web application?

Provide a target URL and testing scope, then follow the five-phase workflow: plan a sitemap, explore pages with browser navigation and snapshots, collect screenshot evidence for each issue, categorize by severity, and generate a structured report.

How to detect silent JavaScript errors during browser testing?

Call browser_console after every page navigation and significant interaction such as form submissions or clicks. Silent JS errors captured this way are among the highest-value findings and should be recorded with reproduction steps.

What browser tools are required for this QA testing workflow?

The workflow requires browser_navigate, browser_snapshot, browser_click, browser_type, browser_scroll, browser_back, browser_press, browser_vision, and browser_console. The browser_vision tool with annotate=true labels interactive elements for targeted clicking.

What happens when the target application URL is unreachable?

The Skill refuses to run and returns an error containing the failing URL, the failure mode such as connection refused, and at least two recovery suggestions. It never fabricates findings or screenshot evidence for an unreachable target.

How are bugs classified in the QA report?

Each issue is assigned a severity (Critical, High, Medium, Low) and a category (Functional, Visual, Accessibility, Console, UX, Content). Duplicates are merged, issues are sorted by severity, and counts feed the executive summary.