今日热榜中文

Use AI to check website buttons and forms

Start with Playwright MCP for one website workflow. Prepare a test URL and expected outcome, copy the first prompt and verify controls, form values and completion states.

Start with Playwright MCP

Choose Playwright MCP when you want to check a button, fill a form and see whether the page changes as expected. It gives an assistant structured page information so it can operate controls step by step.

MCP is a way for an assistant to connect to tools. You need an assistant that supports it. Choose one page and a short sequence, such as entering a search term, pressing Search and checking the results.

Prepare before starting

You do not send passwords or keys to this website; sign-in and tool connections happen in your own assistant and browser.

  1. A test URL you manage. Start your local project first so a browser can reach it.
  2. A specific sequence, test data and expected outcome. If sign-in is needed, use a test account and sign in through your own browser.
  3. Connect Playwright MCP using your assistant’s documented setup. It requires Node.js 18 or newer; Node.js is the local runtime for these tools. Follow official instructions if a browser is missing.

Copy this first prompt

Use Playwright MCP to check [test page URL]. Confirm the page title, click [button label], enter [test value] in [field name], and check for [expected outcome]. Use test data for this workflow. Stop before submission if it sends real messages, makes a payment or changes production data. Record each actual outcome, include the URL and a screenshot for failures, and list any steps you did not run.

How can you verify completion?

  1. Check the tool activity for the URL you provided, and confirm the page title and test account.
  2. Compare each actual outcome with the expected page, field value, error or completion state. Use screenshots as supporting evidence instead of relying on a success claim.
  3. Expect reproducible steps for failures and a separate list of unexecuted steps. When switching tools, reproduce the same failure before inspecting its error or request information.

Actual example

When should you switch to Chrome DevTools MCP?

Choose Chrome DevTools MCP when you can reproduce the problem and need to investigate an error or slow load. It can inspect network requests, console messages and performance recordings. The console is where a page reports runtime errors and messages.

Both tools can operate a page. Switch for the diagnostic information you need; a button-and-form check does not require both tools by default.

For Chrome DevTools MCP, check its Node.js LTS, npm and Chrome requirements. npm installs Node.js tools.

Sources and next step

Sources checked on 2026-10-13. This first prompt is a template based on official documentation; the template itself is not evidence that the workflow has run. Use the resulting records to decide what to fix.

Sources and ranking

This directory lists AI development plugins with public source code and a clear purpose. Browse by task, then open an introduction to learn what a project does, who it is for, and what you need to get started. It covers selected projects, not every AI plugin.

How do the three plugin types differ?

Tool connections (MCP) let AI assistants use external tools such as browsers, code repositories and databases. MCP is the protocol used for these connections.

Reusable skills give a compatible AI assistant instructions and guidance for a particular task.

Editor plugins run inside the software you use to write code, adding AI chat, completions or suggested edits.

How does “Most starred” work?

Projects are sorted by their total GitHub stars. Stars indicate interest; they are not downloads, active users or quality scores. A skill pack counts as one project. Daily star growth is not shown without historical records.

What does “Recently updated” mean?

It uses the most recent code push, which may not be a new release. The 7-day filter checks that project update time; ranking still uses all-time stars. Each introduction shows when its star count was collected. Failed updates keep the last saved data.

Introductions are based on public project documentation. These projects have not each been installed and tested here. Check the project documentation for setup, compatibility and service costs. Refreshing this list only reads saved data; it does not start collection.

UTC+8