home / mcp / mcp server heroku

MCP Server Heroku

Provides an MCP server with both HTTP SSE endpoint and local stdio tooling for Cursor IDE.

Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "myblockcities-mcp-server-heroku": {
      "url": "https://<your-app-name>.herokuapp.com/sse",
      "headers": {
        "DEBUG": "false",
        "MCP_USER_AGENT": "MyMCPServer/1.0",
        "MCP_SERVER_HOST": "0.0.0.0",
        "MCP_SERVER_PORT": "8000"
      }
    }
  }
}

You can run a dedicated MCP server to connect Cursor IDE to external tools using the Model Context Protocol. This server exposes an SSE endpoint for live communication and can also run locally via a standard CLI tool, giving you flexible options for integrating your workflows with Cursor.

How to use

Connect Cursor to your MCP server by pointing the MCP client at the server’s SSE URL. For a remote deployment, use the HTTPS SSE URL provided by your hosting setup (for example, the endpoint ending with /sse). You can test the connection by sending a simple query through the MCP channel and observing the server’s responses in Cursor. If you also run the server locally, you can connect Cursor to http://localhost:8000/sse and interact in the same way.

How to install

Prerequisites you need before starting are Node.js tools for the server, a container runtime if you choose Docker, and access to a shell where you can run commands.

# Docker setup (preferred for quick start)
# Clone the repository
git clone https://github.com/kirill-markin/weaviate-mcp-server.git
cd weaviate-mcp-server

# Create environment file
cp .env.example .env

# Build and run with Docker Compose
docker compose up --build -d

docker compose logs -f

docker compose ps

# Stop the server
docker compose down

# SSE endpoint (local quick test)
curl -i http://localhost:8000/sse

Alternative traditional setup uses the uv package manager to install and run the MCP server locally.

# Install uv and the server in development mode
uv pip install -e ".[dev]"

# Using stdio transport (default)
uv run mcp-simple-tool

# Using SSE transport on a custom port
uv run mcp-simple-tool --transport sse --port 8000

# Run tests
uv run pytest -v

Additional setup notes

If you want to deploy remotely, you can use a hosting service that supports running the MCP server and exposes an SSE endpoint at /sse. You can then connect Cursor to the published URL and begin interacting with the server. For local development, the server exposes the SSE endpoint at http://localhost:8000/sse, which you can test with a simple request.

Security and maintenance

Keep your MCP server up to date with the latest security patches from your chosen runtime. If you expose the server publicly, ensure you use secure transport (HTTPS) and restrict access where appropriate. Monitor logs to catch unauthorized access attempts and rotate any credentials used by the server.

Notes on configuration and environment

The server can be configured via environment variables to customize its behavior. Common options include the port, host binding, and debug mode. If you copy or create an .env file, align variables with your deployment needs.