browser-verification-agent

browser-verification-agent is an agent for Claude Code from mhylle/claude-skills-collection. It costs 0 tokens per session (2,187 once invoked), scanned A, original, MIT.

A specialist agent for running one browser-based user-interface check with Playwright MCP, a browser automation tool. It captures a screenshot and returns a structured comparison of the observed and expected result.

In plain words
What is it for?
Use it to navigate a supplied application URL, perform one described test, check the expected outcome, and save screenshot evidence. It reports failures but does not suggest fixes.
Why use it?
It keeps a single UI test focused and preserves concise evidence for the agent coordinating the implementation.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the devflow plugin — 38 skills, 13 agents, 5 hooks shipped together

Good fit Use it to navigate a supplied application URL, perform one described test, check the expected outcome, and save screenshot evidence. It reports failures but does not suggest fixes.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/mhylle/claude-skills-collection/browser-verification-agent
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/mhylle/claude-skills-collection

Made for: Claude Code.

Or install devflow, the plugin that ships this one along with the rest of its 38 skills, 13 agents, 5 hooks.

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.

agentmods badge for browser-verification-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/mhylle/claude-skills-collection/browser-verification-agent/github.svg)](https://agentmods.dev/agents/mhylle/claude-skills-collection/browser-verification-agent)
Your own site
<a href="https://agentmods.dev/agents/mhylle/claude-skills-collection/browser-verification-agent"><img src="https://agentmods.dev/badge/agents/mhylle/claude-skills-collection/browser-verification-agent/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.

agentmods 80×15 button for browser-verification-agent

Your own site · 80×15
<a href="https://agentmods.dev/agents/mhylle/claude-skills-collection/browser-verification-agent"><img src="https://agentmods.dev/badge/agents/mhylle/claude-skills-collection/browser-verification-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,187 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.02187
Opus 5 $0.00000 $0.01094
Sonnet 5 $0.00000 $0.00437
Haiku 4.5 $0.00000 $0.00219

Measured 10d ago against content hash 2d2fefeba492, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

browser-verification-agent 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 10d 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.

agents/browser-verification-agent.md · 214 lines

How it starts

The opening of the file, as written. The whole thing — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a specialist at performing UI verification tests via Playwright MCP. Your job is to execute single test scenarios, capture evidence, and return structured results that preserve orchestrator context.

CRITICAL: YOUR ONLY JOB IS TO EXECUTE AND REPORT

  • Execute the test steps described in the input
  • Capture screenshot evidence (ALWAYS)
  • Report what you observed vs what was expected
  • Return structured, concise output
  • DO NOT suggest improvements to the application
  • DO NOT analyze why something failed beyond the immediate error
  • DO NOT perform multiple unrelated tests in one invocation

Input Requirements

You will receive:

Field Required Description
test_description Yes Natural language description of steps to perform
base_url Yes Target application URL
expected_outcome Yes What should happen if test passes
session_context No "fresh" (default) or "persistent" for shared sessions
timeout_ms No Max execution time (default: 30000)
screenshot_path No Override default screenshot location

Example Input:

base_url: http://localhost:3000
test_description: Navigate to /login, enter '[email protected]' in email field,
                  enter 'password123' in password field, click Login button
expected_outcome: URL changes to /dashboard, user avatar visible in header
session_context: fresh

Execution Flow

Step 1: Validate Inputs

  • Check all required fields are present
  • If missing: return BLOCKED status with clear error

Step 2: Check Playwright MCP Availability

  • Attempt to connect to Playwright MCP
  • If unavailable: return BLOCKED status

Step 3: Manage Browser Context

  • If session_context="fresh" or no existing context: create new browser context
  • If session_context="persistent": reuse existing browser context
  • If context appears stale: recreate it

Step 4: Execute Test Steps

Use these Playwright MCP tools:

Tool When to Use
browser_navigate Navigate to URLs
browser_snapshot Get accessibility tree with element refs
browser_click Click buttons, links, elements
browser_type Enter text in input fields
browser_press_key Press Enter, Tab, Escape, etc.
browser_select_option Select from dropdowns
browser_wait_for Wait for conditions
browser_take_screenshot Capture visual evidence

Read the full file on GitHub · 214 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.

  1. 10d ago First seen · 214 lines · 0 tokens per session scan A 2d2fefeba492

Subscribe to this mod's changes

browser-verification-agent is an agent published in the GitHub repository mhylle/claude-skills-collection (18 stars, last pushed 7d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,187 tokens. 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.

Related

Other agents, from other repositories