Puppeteer FAQs

Question 1

What is Puppeteer MCP Server?

Accepted Answer

It's a tool that uses Puppeteer to automate browser interactions, allowing you to control Chrome instances programmatically. This includes navigation, screenshot capture, element interaction, and JavaScript execution.

Question 2

How do I connect to an existing Chrome instance?

Accepted Answer

Launch Chrome with remote debugging enabled (using the `--remote-debugging-port` flag). Then, use the `puppeteer_connect_active_tab` tool, specifying the target URL and debug port.

Question 3

What are the key features of this tool?

Accepted Answer

Key features include connecting to active Chrome tabs, preserving existing Chrome instances, taking screenshots, clicking elements, filling forms, and executing JavaScript within the browser context.

Question 4

How do I configure Puppeteer MCP Server for Claude?

Accepted Answer

Add the server configuration to your Claude Desktop App or VSCode Extension MCP settings file (claude_desktop_config.json or cline_mcp_settings.json). The configuration details vary depending on whether you installed via npm, npx, or from source. See the README for specific instructions.

Question 5

Where are the logs stored?

Accepted Answer

Logs are stored in the `logs/` directory. Log files are rotated daily, with a maximum size of 20MB per file and a retention period of 14 days. Old logs are automatically compressed.

Puppeteer

关于

主要功能

使用案例