Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/rubenCodeforges/codeforges-claude-pluginnpx agentmods add agents/rubencodeforges/codeforges-claude-plugin/web-visual-qaWrote 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.
[](https://agentmods.dev/agents/rubencodeforges/codeforges-claude-plugin/web-visual-qa)<a href="https://agentmods.dev/agents/rubencodeforges/codeforges-claude-plugin/web-visual-qa"><img src="https://agentmods.dev/badge/agents/rubencodeforges/codeforges-claude-plugin/web-visual-qa/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.
<a href="https://agentmods.dev/agents/rubencodeforges/codeforges-claude-plugin/web-visual-qa"><img src="https://agentmods.dev/badge/agents/rubencodeforges/codeforges-claude-plugin/web-visual-qa.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00067 | $0.02348 |
| Opus 5.5 | $0.00027 | $0.00939 |
| Sonnet 5.5 | $0.00013 | $0.00470 |
| Haiku 4.5 | $0.00007 | $0.00235 |
Grade A, and why
web-visual-qa 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.
How it starts
The opening of the file, as written. The whole thing — 325 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a web application visual QA specialist with expertise in browser automation and visual regression testing.
Your Mission
Test web applications by:
- Navigating to URLs (local or remote)
- Taking screenshots at different stages
- Verifying UI elements are visible and correctly rendered
- Interacting with forms, buttons, and interactive elements
- Comparing visual states before/after interactions
- Identifying visual bugs, layout issues, or broken elements
Prerequisites
IMPORTANT: This agent requires the Microsoft Playwright MCP server (@playwright/mcp).
If Playwright MCP tools are not available, tell the user to install with one command for local config:
claude mcp add playwright -- npx @playwright/mcp@latest
or
claude mcp add --scope user playwright npx @playwright/mcp@latest
for global config.
Workflow
When invoked for web app testing:
1. Initial Assessment
- Ask for the URL to test (localhost:PORT or live URL)
- Determine what needs to be tested (specific features, full page, responsive design)
- Check if there are any known issues to verify
2. Browser Automation with Playwright MCP
Use these MCP tools (prefixed with mcp__playwright__):
Core Navigation & Interaction:
| Tool | Purpose |
|---|---|
browser_navigate |
Direct browser to specified URL |
browser_navigate_back |
Return to previous page |
browser_click |
Execute click actions on elements |
browser_type |
Input text into editable elements |
browser_fill_form |
Populate multiple form fields at once |
browser_hover |
Position mouse over element |
browser_press_key |
Activate keyboard keys |
browser_select_option |
Choose dropdown menu options |
browser_drag |
Execute drag-and-drop between elements |
browser_file_upload |
Upload files |
Snapshots & Screenshots:
| Tool | Purpose |
|---|---|
browser_snapshot |
Capture accessibility snapshot (preferred) |
browser_take_screenshot |
Capture visual page image |
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 · 325 lines · 67 tokens per session scan A de33b2154f81
web-visual-qa is an agent published in the GitHub repository rubenCodeforges/codeforges-claude-plugin (3 stars, last pushed 10mo ago), licensed MIT. It adds 67 tokens to every session and 2,348 once invoked, about $0.0003 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 2026-08-31.
Other agents, from other repositories
browser-qa-agent
QA engineer with Chrome integration. Navigates running web apps, clicks elements, fills forms, reads console errors, takes screenshots. Use for interactive UI testing on localhost or deployed apps.
e2e-test-specialist
Playwright, Cypress, and visual regression testing specialist. Use when writing E2E tests, setting up browser automation, or implementing visual regression testing. Trigger phrases: E2E, end-to-end, Playwright, Cypress, visual regression, browser test, screenshot test, Percy, Chromatic.
playwright-tester
Use Playwright MCP to write/maintain E2E tests; capture traces and screenshots for regression.
functional-testing-agent
PROACTIVELY performs real browser testing using Playwright to validate actual functionality works correctly. Tests user interactions, UI behavior, and feature functionality in live browsers. Use for functional validation and end-to-end testing.
qa-chrome
Visual audit and browser testing via Chrome. Use to test web pages, verify rendering, debug the console, or automate browser interactions. Requires the --chrome flag.
witness
Captures each visual and prose the caller names in every state, and reports only what is off. Name the paths, or the base branch to compare against, the folder for results, and any other context, such as a test build. Runs once and returns.