#820 · Primary category: AI Agents & Automation

mcp-playwright

Playwright Model Context Protocol Server - Tool to automate Browsers and APIs in Claude Desktop, Cline, Cursor IDE and More 🔌

Project last updated:12/13/25

GitHub Stars

5.6K

Forks

533

Contributors

34

License

MIT

Why we included this project

For teams whose AI agents need to actually touch the web, this is one of the most practical bridges between an LLM and a real browser. It is a Model Context Protocol server wrapping Playwright, so a model in Claude Desktop, Cline, Cursor, or VS Code Copilot can navigate pages, take screenshots, scrape content, run JavaScript, and generate test code through ordinary tool calls. The device-emulation work stands out: you can switch to any of 143 real device profiles (iPhone, iPad, Pixel, Galaxy, desktop) with a plain-language request, which makes cross-device testing far less fiddly. Setup is genuinely low-friction. Browsers install automatically on first use, and you can run it over stdio for desktop clients or as a standalone HTTP server for remote and IDE deployments. Teams doing agent-driven web automation, end-to-end testing, or scraping will find a solid, well-documented foundation here.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category