home / mcp / bitteprotocol mcp server
Provides an MCP endpoint for Bitte AI integrations to access data sources and trigger actions.
Configuration
View docs{
"mcpServers": {
"bitteprotocol-mcp": {
"url": "https://mcp.bitte.ai/sse"
}
}
}The MCP server provides a dedicated bridge that lets clients interact with Bitte AI integrations through the Model Control Protocol. It exposes a scalable endpoint you can connect to from your MCP client to access data sources and perform actions in a controlled, programmatic way.
Connect your MCP client to the Bitte AI MCP server using the designated endpoint. Once connected, you can query data sources, trigger actions, and coordinate interactions through MCP messages. Add the server configuration to your client’s cursor/settings so it can automatically route requests to the Bitte AI MCP endpoint.
Prerequisites you need before installation:
- A JavaScript runtime (Bun is recommended).
Follow these steps to set up the MCP server locally:
# Install Bun if you don’t have it yet
curl -fsSL https://bun.sh/install | bash
# Install dependencies for the MCP server package(s)
bun install
# Build all packages in the monorepo
bun run build
# Start the MCP services
bun run startThe Bitte AI MCP server URL you’ll connect to is: https://mcp.bitte.ai/sse. You can add this to your client cursor/settings to enable MCP routing to Bitte AI.
If you need to run in development mode for active development, use the development start command and keep an eye on logs for debugging. The project supports multiple commands for building, starting, and developing packages. Use the explicit commands shown in your environment for the exact setup you’re using.
To configure the cursor with the Bitte AI MCP server, use this exact configuration snippet in your settings.
json
{
"mcpServers": {
"bitte-ai": {
"url": "https://mcp.bitte.ai/sse"
}
}
}Install dependencies for the MCP project using Bun.
Build all packages in the monorepo.
Start all MCP services.
Run for development mode with live reloads.
Lint and format code according to project standards.
Automatically fix linting/formatting issues where possible.
Run the bitte-ai MCP service in development mode.