Get the FREE Ultimate OpenClaw Setup Guide →

mybuz

mybuz-mcp-server hosts tools that has ability to interact with database to get the data for LLM for it to be able to answer questions about Customers

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio ravikantheragapati-mybuz-mcp-server node server.js \
  --env PORT="9090" \
  --env DB_HOST="localhost" \
  --env DB_NAME="your_db_name" \
  --env DB_PORT="5432" \
  --env DB_USER="your_db_user" \
  --env LOG_LEVEL="info" \
  --env DB_PASSWORD="your_db_password"

How to use

mybuz is an MCP server that hosts tools capable of interacting with a database to fetch data about customers so that an LLM can answer questions with accurate, data-backed responses. The server exposes endpoints for retrieving customer information and related data, which can be used to populate prompts for the model context. You can inspect and test the MCP tooling and data access using the MCP Inspector at http://localhost:6274/ and then query the running server at http://localhost:9090/see to understand the available see-through endpoints. To begin, ensure the server is running and the inspector UI is accessible, then explore the customer data access capabilities and how prompts should be structured to leverage the data reliably.

How to install

Prerequisites:

  • Node.js (LTS) installed on your system
  • Access to a database (as configured by environment variables below)

Installation steps:

  1. Clone the repository hosting the MCP server.
  2. Install dependencies:
    • npm install
  3. Create a .env file (optional if you prefer environment variables) with database and server settings, for example: PORT=9090 DB_HOST=localhost DB_PORT=5432 DB_USER=your_db_user DB_PASSWORD=your_db_password DB_NAME=your_db_name LOG_LEVEL=info
  4. Start the server: node server.js or use npm start if defined in package.json
  5. Open the MCP Inspector to verify tooling: npx @modelcontextprotocol/inspector Then visit http://localhost:6274/ to inspect available tools.

Note: The README mentions an inspector workflow and a see endpoint at http://localhost:9090/see; ensure your database credentials and ports match your environment.

Additional notes

Tips and considerations:

  • The server appears to expose a see-like endpoint at http://localhost:9090/see for data access. If you configure a different port, update the mcp_config accordingly.
  • The MCP Inspector tool (npx @modelcontextprotocol/inspector) helps explore available MCP tooling and endpoints before integrating prompts with the LLM.
  • If you encounter connection issues, verify database connectivity, firewall rules, and that the environment variables (DB_HOST, DB_PORT, DB_USER, DB_PASSWORD, DB_NAME) are correctly set.
  • The readme notes that the provided example is outdated; ensure compatibility with current MCP tooling and adjust endpoints or payload formats as needed when integrating with your LLM prompts.
  • You can customize logging via LOG_LEVEL to aid debugging during development.

Related MCP Servers

Sponsor this space

Reach thousands of developers