Cloudflare Browser Rendering Experiments & MCP Server
@amotivv
This project demonstrates how to use Cloudflare Browser Rendering to extract web content for LLM context. It includes experiments with the REST API and Workers Binding API, as well as an MCP server implementation that can be used to provide web context to LLMs.
Overview
What is Cloudflare Browser Rendering Experiments & MCP Server?
This project demonstrates how to use Cloudflare Browser Rendering to extract web content for LLM context. It includes experiments with the REST API and Workers Binding API, as well as an MCP server implementation that provides web context to LLMs via tools like fetching pages and summarizing content.
How to use Cloudflare Browser Rendering Experiments & MCP Server?
Clone the repository, install dependencies, and configure a Cloudflare Worker with the Browser Rendering binding and the nodejs_compat flag. Deploy the Worker, set the BROWSER_RENDERING_API environment variable to your Worker’s URL, then build and run the MCP server with npm run build and npm start. The MCP server exposes tools that can be invoked through compatible clients (e.g., Cline via MCP settings).
Key features of Cloudflare Browser Rendering Experiments & MCP Server
- Demonstrates Cloudflare Browser Rendering for LLM context
- Includes REST API and Workers Binding API experiments
- Provides an MCP server with four tools
- Tools: fetch_page, search_documentation, extract_structured_content, summarize_content
- Integrates with Cline via MCP settings
Use cases of Cloudflare Browser Rendering Experiments & MCP Server
- Extract web page content for LLM context
- Search and retrieve Cloudflare documentation
- Extract structured content from pages using CSS selectors
- Summarize web content for concise LLM input
FAQ from Cloudflare Browser Rendering Experiments & MCP Server
What prerequisites are needed?
Node.js v16 or later, a Cloudflare account with Browser Rendering enabled, TypeScript, and the Wrangler CLI.
How do I configure the Cloudflare Worker?
Set up wrangler.toml with a [browser] binding (e.g., binding = "browser"), enable nodejs_compat, then deploy with npx wrangler deploy.
How do I set the Browser Rendering API URL?
Set the BROWSER_RENDERING_API environment variable to the URL of your deployed Cloudflare Worker.
What MCP server tools are available?
The MCP server provides fetch_page, search_documentation, extract_structured_content, and summarize_content.
What are key learnings for using Cloudflare Browser Rendering?
Use the @cloudflare/puppeteer package, launch the browser with puppeteer.launch(env.browser), enable the nodejs_compat compatibility flag, and always close the
