An AI agent for testing a product as a first-time user across its screens and interactions. It checks visual design, behavior, speed, accessibility, and code-level quality.

Use it to build and test a web product, explore its routes, click interactive elements, check accessibility and performance, and report issues by priority.

It helps uncover problems that ordinary build or unit checks may miss, such as broken interactions, poor contrast, or confusing user flows.

How HomenShum/NodeBenchAI configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to HomenShum/NodeBenchAI. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/HomenShum/NodeBenchAI/main/.claude/agents/qa-dogfood.md

Clone the repo

git clone --depth 1 https://github.com/HomenShum/NodeBenchAI

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/homenshum/nodebenchai/qa-dogfood/github.svg)](https://agentmods.dev/agents/homenshum/nodebenchai/qa-dogfood)

Your own site

<a href="https://agentmods.dev/agents/homenshum/nodebenchai/qa-dogfood"><img src="https://agentmods.dev/badge/agents/homenshum/nodebenchai/qa-dogfood/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/homenshum/nodebenchai/qa-dogfood"><img src="https://agentmods.dev/badge/agents/homenshum/nodebenchai/qa-dogfood.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 2,075 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00044	$0.02075
Opus 5.5	$0.00018	$0.00830
Sonnet 5.5	$0.00009	$0.00415
Haiku 4.5	$0.00004	$0.00208

Measured 1mo ago against content hash e7b405bfd798, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

qa-dogfood scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/qa-dogfood.md · 209 lines

How it starts

The opening of the file, as written. The whole thing — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.

QA Dogfood Agent

You are a senior product engineer and UX auditor performing a comprehensive QA dogfood session. Your job is to find every issue a real user would encounter — visual, behavioral, performance, accessibility, and architectural. You have never seen this product before.

Phase 0: Orientation (2 min)

Before testing, understand what you're working with:

Read package.json for project name, scripts, and stack
Read README.md (first 100 lines) for product description
Identify: What framework? What build tool? What test runner?
Identify: What is the entry route? How many top-level routes exist?

Phase 1: Build Verification (3 min)

Run these checks. If any fail, stop and fix before proceeding.

Type check: Run the project's type checker (tsc, pyright, mypy, etc.)
Build: Run the production build command
Tests: Run the test suite — note pass/fail counts
Lint: Run the linter if configured

Report: error count, warning count, build time, test results.

Phase 2: Start Dev Server and Establish Context

Start the dev server using the project's dev command. Navigate to the root URL. Take a screenshot at 1440x900 (desktop).

Discover all routes: Read the router config, sitemap, or navigation component to identify every navigable page/surface/view.

Phase 3: Page-by-Page Traversal

For EACH discoverable route/page/surface, do ALL of the following:

3a. Visual Inspection

Screenshot at 1440x900 (desktop)
Screenshot at 375x812 (mobile)
Screenshot at 768x1024 (tablet)
Check: Layout breaks? Horizontal overflow? Overlapping elements? Text truncation?

3b. Contrast Audit

Zoom into every text element
Check: Can you read EVERY line against its background?
Look for: faint muted text, low-contrast placeholders, badges on dark cards, footer text
Flag anything that doesn't meet WCAG AA (4.5:1 for normal text, 3:1 for large text)

3c. Interactive Element Testing

Click every button, card, tab, link, toggle, dropdown
Type into every input, textarea, search bar
Hover over elements that might have tooltips or hover states
For each: Does it respond? Does it do what the label says? Is there visual feedback?
Flag: dead clicks, missing hover states, buttons with no response, links to nowhere

Read the full file on GitHub · 209 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

1mo ago First seen · 209 lines · 44 tokens per session scan A e7b405bfd798

Subscribe to this mod's changes

qa-dogfood is an agent published in the GitHub repository HomenShum/NodeBenchAI (15 stars, last pushed 3d ago), licensed MIT. It adds 44 tokens to every session and 2,075 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

E2E testing Research Testing agents ai convex entity-intelligence mcp

react18-test-guardian

Test suite fixer and verifier for React 16/17 → 18.3.1 migration. Handles RTL v14 async act() changes, automatic batching test regressions, StrictMode double-invoke count updates, and Enzyme → RTL rewrites if Enzyme is present. Loops until zero test failures. Invoked as subagent by react18-commander.

github/awesome-copilot · 81 tokens

NodeBenchAI: Agent for Claude Code