Playwrightを用いたブラウザ自動化機能を提供するモデルコンテキストプロトコル（LLM）サーバー。このサーバーにより、LLMはWebページとのインタラクション、スクリーンショットの取得、テストコードの生成、ページのWebスクレイピング、そして実際のブラウザ環境でのJavaScript実行が可能になります。
Related MCP server: Puppeteer MCP Server
npm、mcp-get、Smithery のいずれかを使用してパッケージをインストールできます。
npm を使用する場合:
npm install -g @executeautomation/playwright-mcp-servermcp-get の使用:
npx @michaellatman/mcp-get@latest install @executeautomation/playwright-mcp-server鍛冶屋を使う
Smithery経由で Claude Desktop 用の Playwright MCP を自動的にインストールするには:
npx @smithery/cli install @executeautomation/playwright-mcp-server --client claude次のいずれかのボタンを使用して、VS Code に Playwright MCP サーバーをインストールします。
あるいは、VS Code CLI を使用して Playwright MCP サーバーをインストールすることもできます。
# For VS Code
code --add-mcp '{"name":"playwright","command":"npx","args":["@executeautomation/playwright-mcp-server"]}'# For VS Code Insiders
code-insiders --add-mcp '{"name":"playwright","command":"npx","args":["@executeautomation/playwright-mcp-server"]}'インストール後、ExecuteAutomation Playwright MCP サーバーを VS Code の GitHub Copilot エージェントで使用できるようになります。
Playwright サーバーを使用するための Claude Desktop の構成は次のとおりです。
{
"mcpServers": {
"playwright": {
"command": "npx",
"args": ["-y", "@executeautomation/playwright-mcp-server"]
}
}
}このプロジェクトではテストにJestを使用しています。テストはsrc/__tests__ディレクトリにあります。
次のいずれかのコマンドを使用してテストを実行できます。
# Run tests using the custom script (with coverage)
node run-tests.cjs
# Run tests using npm scripts
npm test # Run tests without coverage
npm run test:coverage # Run tests with coverage
npm run test:custom # Run tests with custom script (same as node run-tests.cjs)テスト カバレッジ レポートは、 coverageディレクトリに生成されます。
evalsパッケージはmcpクライアントをロードし、index.tsファイルを実行するため、テスト間で再構築する必要はありません。npxコマンドの先頭に環境変数をロードすることもできます。完全なドキュメントはこちらでご覧いただけます。
OPENAI_API_KEY=your-key npx mcp-eval src/evals/evals.ts src/tools/codegen/index.tsClear a code generation session without generating a test
Inputs: sessionId (required).
End a code generation session and generate the test file
Inputs: sessionId (required).
Get information about a code generation session
Inputs: sessionId (required).
Wait for and validate a previously initiated HTTP response wait operation.
Inputs: id (required), value.
Click an element on the page
Inputs: selector (required).
Click a link and switch to the newly opened tab
Inputs: selector (required).
Close the browser and release all resources
Inputs: no named parameters.
Retrieve console logs from the browser with filtering options
Inputs: clear, limit, search, type.
Set a custom User Agent for the browser
Inputs: userAgent (required).
Perform an HTTP DELETE request
Inputs: url (required).
Drag an element to a target location
Inputs: sourceSelector (required), targetSelector (required).
Execute JavaScript in the browser console
Inputs: script (required).
Ask Playwright to start waiting for a HTTP response. This tool initiates the wait operation but does not wait for its completion.
Inputs: id (required), url (required).
fill out an input field
Inputs: selector (required), value (required).
Perform an HTTP GET request
Inputs: url (required).
Get the HTML content of the current page
Inputs: no named parameters.
Get the visible text content of the current page
Inputs: no named parameters.
Navigate back in browser history
Inputs: no named parameters.
Navigate forward in browser history
Inputs: no named parameters.
Hover an element on the page
Inputs: selector (required).
Click an element in an iframe on the page
Inputs: iframeSelector (required), selector (required).
Fill an element in an iframe on the page
Inputs: iframeSelector (required), selector (required), value (required).
Perform an HTTP PATCH request
Inputs: url (required), value (required).
Perform an HTTP POST request
Inputs: url (required), value (required), headers, token.
Press a keyboard key
Inputs: key (required), selector.
Perform an HTTP PUT request
Inputs: url (required), value (required).
Save the current page as a PDF file
Inputs: outputPath (required), filename, format, margin, +1 more.
Take a screenshot of the current page or a specific element
Inputs: name (required), downloadsDir, fullPage, height, +4 more.
Select an element on the page with Select tag
Inputs: selector (required), value (required).
Start a new code generation session to record Playwright actions
Inputs: options (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 31 tools
Most tools have distinct purposes, with clear separation between browser actions (e.g., click, fill, navigate), HTTP methods (get, post, put), and code generation sessions. However, some overlap exists between playwright_click and playwright_iframe_click, where the iframe variant is a specialized case rather than a fundamentally different operation, which could cause minor confusion.
Tool names follow a highly consistent snake_case pattern with clear verb_noun structures. Code generation tools use consistent 'codegen_session' terminology, while Playwright actions consistently start with 'playwright_' followed by descriptive verbs (e.g., playwright_click, playwright_navigate). This predictability makes the set easy to navigate and understand.
With 31 tools, the count feels heavy for a Playwright automation server, suggesting potential over-fragmentation. While Playwright is feature-rich, many tools could be consolidated (e.g., separate HTTP methods might be unified into a single playwright_request tool). The number is borderline excessive for typical agent workflows.
The toolset provides comprehensive coverage for Playwright automation, including navigation, interaction (click, fill, hover), HTTP requests, iframe handling, debugging (logs, screenshots), and code generation. It supports full CRUD-like operations for web testing without obvious gaps, enabling agents to handle complex browser automation tasks end-to-end.
Official Browserless MCP server: browser automation, scraping, screenshots and PDFs for AI agents.
Web scraping, browser automation, screenshots and AI data extraction for AI agents via MCP.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Cloud browser, web extraction, screenshots, and remote CDP tools for AI agents.