GitHubスター
0
ユーザー評価
未評価
お気に入り
0
閲覧数
5
フォーク
0
イシュー
0
README
Browser Automation MCP
This MCP (Model Context Protocol) server provides tools for browser automation using Puppeteer.
Features
- Launch Browser: Launches a new browser instance (Chromium by default).
- Close Browser: Closes an active browser instance.
- New Page: Opens a new page/tab in an active browser.
- Go To URL: Navigates a page to a specified URL.
- Get Content: Retrieves the full HTML content of a page.
- Take Screenshot: Captures a screenshot of the current page.
- Click Element: Clicks on an element specified by a CSS selector.
- Type Text: Types text into an element specified by a CSS selector.
- Wait For Selector: Waits for an element matching a CSS selector to appear on the page.
- Execute JavaScript: Executes arbitrary JavaScript code in the page context.
Setup
- Install dependencies:
npm install(this will also download Chromium for Puppeteer) - Compile TypeScript:
npm run build - Run the server:
node build/index.js
Known Issues
- Tool Listing: The server currently fails to properly list its available tools via the standard MCP discovery mechanism. This is a known issue and will be addressed in a future update.
- Error Handling: Robust error handling for various Puppeteer scenarios is still under development.
- Resource Management: Ensure browser instances are properly closed to avoid resource leaks, especially in long-running applications.