mcp
MCP server from divslingerx/mcp-server
claude mcp add --transport stdio divslingerx-mcp-server node dist/server.js \ --env PORT="3000" \ --env PUPPETEER_HEADLESS="true" \ --env PUPPETEER_EXECUTABLE_PATH="/path/to/chrome"
How to use
This MCP server provides a web search capability powered by Puppeteer. It exposes a tool named search_web that MCP clients can invoke to perform Google searches and receive structured JSON results. The server runs locally (or in your environment) and uses a headless Chromium instance to render search results, then returns a machine-friendly JSON payload with relevant snippets, titles, URLs, and cached metadata. To use it, connect to the MCP hub and call the search_web tool with a query parameter. For example, sending a request with tool set to search_web and arguments containing a query like i) {"tool": "search_web", "arguments": {"query": "example search"}} will return a structured response containing the top results. The server is designed to be lightweight and stateless, making it easy to scale and integrate with MCP-enabled systems.
How to install
Prerequisites:
- Node.js (v14+ recommended)
- Git
- npm (comes with Node.js)
Installation steps:
-
Clone the repository git clone https://github.com/yourusername/mcp-server.git cd mcp-server
-
Install dependencies npm install
-
Build the project npm run build
-
Create an environment file (optional but recommended)
- Create a .env file in the project root and configure Puppeteer and server settings
- Start the server npm start
Notes:
- Ensure you have a compatible Chromium/Chrome path if using a custom PUPPETEER_EXECUTABLE_PATH
- Adjust PORT in .env or environment variables if needed
Additional notes
Tips and common issues:
- Puppeteer may require a compatible Chrome/Chromium binary. If using a custom path, set PUPPETEER_EXECUTABLE_PATH accordingly.
- In headless mode, some environments (like certain CI systems) may require additional flags (e.g., --no-sandbox, --disable-setuid-sandbox). Add them to your Puppeteer launch configuration if needed.
- The server exposes the search_web tool via MCP. If you see timeouts, verify that the port is open and that the MCP hub can reach the server.
- Environment variables can be overridden at runtime. Use a .env file or export VAR=value in the hosting environment.
- Build step outputs to dist/server.js (as assumed). If your build outputs elsewhere, adjust the mcp_config accordingly.
Related MCP Servers
zen
Selfhosted notes app. Single golang binary, notes stored as markdown within SQLite, full-text search, very low resource usage
MCP -Deepseek_R1
A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)
mcp-fhir
A Model Context Protocol implementation for FHIR
mcp
Inkdrop Model Context Protocol Server
mcp-appium-gestures
This is a Model Context Protocol (MCP) server providing resources and tools for Appium mobile gestures using Actions API..
dubco -npm
The (Unofficial) dubco-mcp-server enables AI assistants to manage Dub.co short links via the Model Context Protocol. It provides three MCP tools: create_link for generating new short URLs, update_link for modifying existing links, and delete_link for removing short links.