agentmods.dev

browser-qa

browser-qa is an agent for Claude Code from VKirill/claude-lane-stack. It costs 67 tokens per session (941 once invoked), scanned A, original, MIT.

A browser-testing agent that opens a supplied URL, follows click-based acceptance scenarios, and saves replay scripts, screenshots, and a report. It uses a live browser rather than fetching page text.

In plain words
What is it for?
Use it for live acceptance testing at viewports such as 375, 768, and 1280 pixels, replaying scenarios, checking staging or preview pages, and sending back failure and screenshot paths.
Why use it?
It provides evidence about what users see and can do, including failures caused by missing controls or incorrect layouts. It keeps browser checks separate from product code and design reviews.

Agent for Claude Code

Written for Claude Code: background in frontmatter. Also seen: model in frontmatter; mentions Codex.

Part of the lane-stack plugin — 62 skills, 4 commands, 13 agents, 1 MCP server shipped together

not rated 122repo changed 26d ago A scan 67 tokens original MIT

Good fit Use it for live acceptance testing at viewports such as 375, 768, and 1280 pixels, replaying scenarios, checking staging or preview pages, and sending back failure and screenshot paths.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/vkirill/claude-lane-stack/browser-qa
View source ↗ VKirill/claude-lane-stack
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/VKirill/claude-lane-stack

Made for: Claude Code.

Or install lane-stack, the plugin that ships this one along with the rest of its 62 skills, 4 commands, 13 agents, 1 MCP server.

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-qa

README.md
[![agentmods](https://agentmods.dev/badge/agents/vkirill/claude-lane-stack/browser-qa/github.svg)](https://agentmods.dev/agents/vkirill/claude-lane-stack/browser-qa)
Your own site
<a href="https://agentmods.dev/agents/vkirill/claude-lane-stack/browser-qa"><img src="https://agentmods.dev/badge/agents/vkirill/claude-lane-stack/browser-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.

agentmods 80×15 button for browser-qa

Your own site · 80×15
<a href="https://agentmods.dev/agents/vkirill/claude-lane-stack/browser-qa"><img src="https://agentmods.dev/badge/agents/vkirill/claude-lane-stack/browser-qa.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 67 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 941 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 $0.00067 $0.00941
Opus 5.5 $0.00027 $0.00376
Sonnet 5.5 $0.00013 $0.00188
Haiku 4.5 $0.00007 $0.00094

Measured 26d ago against content hash 55f94baaf34f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

browser-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 26d 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.

plugins/lane-stack/agents/browser-qa.md · 49 lines

How it starts

The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.

browser-qa

You collect browser evidence. You are not the PM and not a writer.

Model

Read stages.browser_qa from .agents/routing.profile.yaml (adoc → Stages). Default provider: jev, backend: chrome-qa: browser-qa-jev snapshots visible controls and Jev picks the next click. provider: codex is Astra + browser plugins. provider: claude means you drive chrome-devtools MCP yourself.

Claude Haiku + chrome-devtools MCP (provider claude). Host-level server mcp__chrome-devtools__* first (desktop hosts: chrome-qa start profile on --browserUrl http://127.0.0.1:9333, or --autoConnect to the user's real Chrome); MetaMCP mcp_call server chrome-devtools as fallback. Not Codex. Not Fable.

Inputs

PROJECT_CWD. Required: URL or BASE_URL. Optional: SLUG, CASES (acceptance bullets), SPEC path, VIEWPORTS (default 375,768,1280), ENV_CLASS (local/staging/preview).

Run

  1. Load skill browser-qa. Do not dump it.
  2. cd "$PROJECT_CWD". English files only under .agents/qa/<slug>/. 2a. provider jev (default): run once via Bash browser-qa-jev --project-cwd "$PROJECT_CWD" --url "$URL" --slug "$SLUG" --env-class "$ENV_CLASS" --viewports "$VIEWPORTS" --case "<bullet>" ... (add --authorized only when the PM wrote it). Writes cases.md, REPORT.md, shots/, jev-run.json. Exit 0 pass · 1 fail · 2 blocked. Then step 6. 2b. provider codex: browser-qa-codex with the same flags. Writes codex-run.json
    • codex-result.json. Then step 6. Do not click yourself.
  3. Follow skill browser-qa (AI Factory #154 adapted): qa-digest, stale rules, one script per case, first run is proof, missing selector is a product fail, no pass without a live browser. Use references/ templates. Write secrets-free .agents/qa/context.md only for reusable setup.
  4. Browser tools, in order: host-level mcp__chrome-devtools__list_pages / new_page / navigate_page → take_snapshot → clicks → take_screenshot; else MetaMCP mcp_call server chrome-devtools. Not WebFetch. If the MCP has no browser: chrome-qa start, retry once. When attached to the user's real Chrome (--autoConnect): open your own page with new_page, work only in pages you created, never read or close the user's other tabs, close yours at the end.
  5. No Vue/TS/CSS. No docs/DESIGN.md. Taste/slop → tell PM to use design-lead MODE=audit.
  6. Last line: DONE .agents/qa/<slug>/REPORT.md or FAILED <reason>, then stop.

Read the full file on GitHub · 49 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. 26d ago Changed 55f94baaf34f
  2. 1mo ago Changed · +11 lines 8035ecf0ee32
  3. 1mo ago First seen · 38 lines · 67 tokens per session scan A 67d964735b11

Subscribe to this mod's changes

browser-qa is an agent published in the GitHub repository VKirill/claude-lane-stack (122 stars, last pushed 2d ago), licensed MIT. It adds 67 tokens to every session and 941 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 2026-09-14.

Browser automationClaude CodeE2E testingTestingagentic-codingai-agentsai-codingautomationclaude
Related

Other agents, from other repositories

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.

travisjneuman/.claude · 67 tokens

orc-qa-validator

Executor role — drives a real browser via the agent-browser CLI (vercel-labs/agent-browser) to QA a running web application — golden path + edge cases — and writes an evidence packet (criterion-anchored annotated screenshots, a WebM recording, accessibility snapshots, browser console, network HAR, narrated steps…

HigorAlves/orc · 110 tokens

playwright-e2e

Playwright E2E testing specialist. Use for writing, reviewing, debugging, and optimizing Playwright tests. Proactively use when working with test files (.spec.ts), page objects, or test helpers.

ivklgn/ai-kit · 47 tokens

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.

christopherlouet/claude-base · 36 tokens

integration-tester

Independent dynamic verification agent. Runs the test suite, executes acceptance criteria commands, and verifies runtime behavior. Breaks the self-assessment cycle where the implementer verifies their own work. Reports with command output evidence.

joris887/exosuit · 48 tokens

smoke-sonnet

Use when one functional smoke flow must be walked in the running app as one persona, with a pass/fail per step. Needs the chrome-devtools MCP server. Handoff - base URL, persona credentials, the numbered steps with what must hold at each, the viewport, and a screenshot directory. Returns a step table; visual doubts…

yacb2/aidex · 105 tokens