ワンクリックでManusで任意のスキルを実行

$pwd:

qa-browser-verify

Name: Qa Browser Verify
Author: hoangnb24

// Browser-first QA skill for local web apps. Use when exploring an application, smoke-testing routes, checking whether a page loads, validating a user flow, reproducing a UI bug, or confirming a fix. Prefer the local `agent-browser` CLI for screenshots, snapshots, console checks, navigation, and basic interactions.

Manusで実行

$ git log --oneline --stat

stars:5

forks:5

updated:2026年4月16日 08:07

ファイルエクスプローラー

2 ファイル

SKILL.md

readonly

name	qa-browser-verify
description	Browser-first QA skill for local web apps. Use when exploring an application, smoke-testing routes, checking whether a page loads, validating a user flow, reproducing a UI bug, or confirming a fix. Prefer the local `agent-browser` CLI for screenshots, snapshots, console checks, navigation, and basic interactions.

QA Browser Verify

Use this skill for manual QA passes against a running local application.

Default stance

Prefer agent-browser over curl for UI verification.
Verify what the user actually sees: rendered content, overlays, console errors, and basic interactions.
Stay in QA mode: gather evidence first, describe failures clearly, and avoid editing code unless the user asks for a fix.
Run agent-browser commands sequentially for a single session. Do not fire multiple commands at the same session in parallel.

Setup

Identify the dev server URL from the repo or running processes.
If agent-browser is missing, stop and report that blocker.
If browser launch fails because Playwright browsers are missing, run:

agent-browser install
npx playwright install chromium chromium-headless-shell

Use a dedicated session name, for example:

AGENT_BROWSER_SESSION=qa agent-browser open http://localhost:3000

Smoke-test flow

Run this loop for the home route and any important feature routes:

AGENT_BROWSER_SESSION=qa agent-browser open http://localhost:3000
AGENT_BROWSER_SESSION=qa agent-browser screenshot /tmp/qa-home.png
AGENT_BROWSER_SESSION=qa agent-browser snapshot -i
AGENT_BROWSER_SESSION=qa agent-browser eval 'document.querySelector("[data-nextjs-dialog], .vite-error-overlay, #webpack-dev-server-client-overlay") ? "ERROR_OVERLAY" : "OK"'
AGENT_BROWSER_SESSION=qa agent-browser eval 'document.body.innerText.trim().length > 0 ? "HAS_CONTENT" : "BLANK"'
AGENT_BROWSER_SESSION=qa agent-browser console
AGENT_BROWSER_SESSION=qa agent-browser errors

Interaction checks

Use fill, type, click, press, and get url to validate a real user path.
After submitting a form or triggering a flow, wait briefly and capture a second screenshot.
When useful, confirm the page content with get text, eval, or another snapshot -i.

Example:

AGENT_BROWSER_SESSION=qa agent-browser fill 'input[name="prompt"]' 'Hello'
AGENT_BROWSER_SESSION=qa agent-browser click 'button[type="submit"]'
AGENT_BROWSER_SESSION=qa agent-browser wait 3000
AGENT_BROWSER_SESSION=qa agent-browser screenshot /tmp/qa-after-submit.png

Failure capture

When a check fails:

Save a screenshot of the broken state.
Capture console and errors.
If the request flow matters, inspect network requests.
Report the exact route, user action, observed result, and expected result.

Closeout

Finish by closing the browser session:

AGENT_BROWSER_SESSION=qa agent-browser close

Return a concise QA summary:

what routes were checked
what interactions were tested
pass/fail status
concrete evidence for any failure

related-skills.json

同じリポジトリ

agent-creator.md

from "hoangnb24/ai20k-demo"

Create new persistent Codex agents for this repo when asked to add a specialist agent, define a reusable subagent role, scaffold `.codex/agents/*.toml`, or wire the agent to local skills and MCP servers.

2026-04-165

vercel-nextjs-frontend.md

from "hoangnb24/ai20k-demo"

Repo-local bridge skill for Next.js frontend work. Use when a dedicated agent is handling App Router UI changes, layouts, metadata, route-aware rendering, or Server/Client Component boundaries.

2026-04-165

vercel-react-best-practices.md

from "hoangnb24/ai20k-demo"

Repo-local bridge skill for React TSX reviews. Use when a dedicated agent is checking component quality, hooks usage, accessibility, and performance risks after frontend changes.

2026-04-165

vercel-shadcn-ui.md

from "hoangnb24/ai20k-demo"

Repo-local bridge skill for shadcn/ui work. Use when a dedicated agent is composing UI, extending the design system, or adding shadcn components in a Next.js frontend.

2026-04-165

stack-scout.md

from "hoangnb24/ai20k-demo"

Research the current best implementation pattern before making code changes. Use when Codex needs up-to-date guidance for a feature, integration, bugfix, or upgrade, especially around Next.js, AI SDK, Gemini, Drizzle, Bun, Docker, or other stack pieces that may have changed recently.

2026-04-165

postgres-seed-lab.md

from "hoangnb24/ai20k-demo"

Turn plain-language requests into transient query or mutation scripts for the local seeded Postgres database in this repo. Use when Codex needs to inspect, clean up, backfill, bulk update, or analyze demo_orders or other local seed data, preferably through DOCKER MCP code-mode and otherwise through js_repl without requiring a checked-in script.

2026-04-165

package.json

"author": "hoangnb24"

"repository": "hoangnb24/ai20k-demo"

GitHub リポジトリを開く Creator のリポジトリを見る

$ install --global

$ download --local

Manusで実行

$ useful --forSOC

ソフトウェア品質保証アナリスト・テスターコンピュータ・数学職15-1253L4

name	qa-browser-verify
description	Browser-first QA skill for local web apps. Use when exploring an application, smoke-testing routes, checking whether a page loads, validating a user flow, reproducing a UI bug, or confirming a fix. Prefer the local `agent-browser` CLI for screenshots, snapshots, console checks, navigation, and basic interactions.

QA Browser Verify

Use this skill for manual QA passes against a running local application.

Default stance

Prefer agent-browser over curl for UI verification.
Verify what the user actually sees: rendered content, overlays, console errors, and basic interactions.
Stay in QA mode: gather evidence first, describe failures clearly, and avoid editing code unless the user asks for a fix.
Run agent-browser commands sequentially for a single session. Do not fire multiple commands at the same session in parallel.

Setup

Identify the dev server URL from the repo or running processes.
If agent-browser is missing, stop and report that blocker.
If browser launch fails because Playwright browsers are missing, run:

agent-browser install
npx playwright install chromium chromium-headless-shell

Use a dedicated session name, for example:

AGENT_BROWSER_SESSION=qa agent-browser open http://localhost:3000

Smoke-test flow

Run this loop for the home route and any important feature routes:

AGENT_BROWSER_SESSION=qa agent-browser open http://localhost:3000
AGENT_BROWSER_SESSION=qa agent-browser screenshot /tmp/qa-home.png
AGENT_BROWSER_SESSION=qa agent-browser snapshot -i
AGENT_BROWSER_SESSION=qa agent-browser eval 'document.querySelector("[data-nextjs-dialog], .vite-error-overlay, #webpack-dev-server-client-overlay") ? "ERROR_OVERLAY" : "OK"'
AGENT_BROWSER_SESSION=qa agent-browser eval 'document.body.innerText.trim().length > 0 ? "HAS_CONTENT" : "BLANK"'
AGENT_BROWSER_SESSION=qa agent-browser console
AGENT_BROWSER_SESSION=qa agent-browser errors

Interaction checks

Use fill, type, click, press, and get url to validate a real user path.
After submitting a form or triggering a flow, wait briefly and capture a second screenshot.
When useful, confirm the page content with get text, eval, or another snapshot -i.

Example:

AGENT_BROWSER_SESSION=qa agent-browser fill 'input[name="prompt"]' 'Hello'
AGENT_BROWSER_SESSION=qa agent-browser click 'button[type="submit"]'
AGENT_BROWSER_SESSION=qa agent-browser wait 3000
AGENT_BROWSER_SESSION=qa agent-browser screenshot /tmp/qa-after-submit.png

Failure capture

When a check fails:

Save a screenshot of the broken state.
Capture console and errors.
If the request flow matters, inspect network requests.
Report the exact route, user action, observed result, and expected result.

Closeout

Finish by closing the browser session:

AGENT_BROWSER_SESSION=qa agent-browser close

Return a concise QA summary:

what routes were checked
what interactions were tested
pass/fail status
concrete evidence for any failure

qa-browser-verify

QA Browser Verify

Default stance

Setup

Smoke-test flow

Interaction checks

Failure capture

Closeout

このリポジトリの他の Skills

このリポジトリの他の Skills

QA Browser Verify

Default stance

Setup

Smoke-test flow

Interaction checks

Failure capture

Closeout