qa-web

Automate exploratory QA for web UIs with agent-browser across multi-phase testing.

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/cjohnhanson/codelikecody --skill qa-web-cjohnhanson
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: qa-web
Source: https://github.com/cjohnhanson/codelikecody/tree/main/skills/qa-web
Command: npx skills add https://github.com/cjohnhanson/codelikecody --skill qa-web-cjohnhanson

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Exploratory QA for web applications can be time-consuming and brittle when manual checks miss edge cases across viewports and interactions. This Skill provides an agent-driven workflow to systematically survey, check, evaluate, and report on web UI quality.

Core Features & Use Cases

  • Phase 1 — Survey (SFDPOT): Open the target in agent-browser and map coverage across structure, function, data, interfaces, platform, operations, and time to generate a prioritized risk-based checklist.
  • Phase 2 — Check (sub-agents): Dispatch independent checkers that execute specific flows with screenshots and pass/fail reporting for each step.
  • Phase 3 — Evaluate (oracles): Apply FEW HICCUPPS consistency oracles to validate results against documentation, user needs, and real-world expectations.
  • Phase 4 — Fresh eyes: Run a context-free exploration to surface gestalt issues and trust signals.

Quick Start

Launch agent-browser against your web app and run the full QA phases to surface and document issues.

Frequently Asked Questions about qa-web

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

FAQPage Schema
How do I automate exploratory web UI testing across multiple viewports?▼

Automated exploratory web UI testing uses an agent-browser to systematically survey, check, and evaluate application quality across multiple viewports. It dispatches sub-agents to execute specific interaction patterns and capture screenshots for pass/fail reporting.

What is the best way to identify coverage gaps in pre-release web applications?▼

Identifying coverage gaps in pre-release web applications involves mapping structure, function, data, and platform operations using the SFDPOT method. This generates a prioritized, risk-based checklist to systematically validate pre-release verification targets.

How do I apply consistency oracles to validate web UI behavior?▼

Applying consistency oracles validates web UI behavior by checking results against the FEW HICCUPPS framework. This evaluates whether application outputs remain consistent with documentation, user needs, and real-world expectations during exploratory QA.

Does automated web UI testing work for post-change validation without manual checks?▼

Automated web UI testing works for post-change validation by running structured, multi-phase testing workflows. It dispatches independent checkers and runs fresh-eyes context-free exploration to surface gestalt issues and trust signals without manual checks.

What limitations exist when using an agent-browser for continuous web QA?▼

Limitations when using an agent-browser for continuous web QA include the reliance on structured, multi-phase testing workflows that require passing through survey, check, evaluation, fresh-eyes, and reporting phases to properly document issues and consider accessibility.