Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Implementation:Datajuicer Data juicer MCP Server

From Leeroopedia
Knowledge Sources
Domains Tooling
Last Updated 2026-02-14 16:00 GMT

Overview

Concrete tool for the unified CLI entry point of the Data-Juicer MCP server provided by Data-Juicer.

Description

mcp_server is the CLI entry point (dj-mcp) for launching the Data-Juicer MCP server in either "granular-ops" mode (individual operator tools) or "recipe-flow" mode (high-level recipe workflow tools). It parses command-line arguments for mode selection, transport protocol (stdio, sse, streamable-http), and port number. Based on the selected mode, it imports the appropriate create_mcp_server function from either DJ_mcp_granular_ops or DJ_mcp_recipe_flow, creates the server, and starts it with the specified transport.

Usage

Use when you need to start Data-Juicer as an MCP-compatible tool server for AI assistant integration, choosing between granular operator access or recipe-based workflow modes.

Code Reference

Source Location

Signature

def main():
    """Data-Juicer MCP Server CLI entry point."""

Import

from data_juicer.tools.mcp_server import main

I/O Contract

Inputs

Name Type Required Description
mode str (CLI arg) Yes MCP server mode: "granular-ops" or "recipe-flow"
--transport str (CLI arg) No Transport protocol: "stdio", "sse", or "streamable-http". Default: "stdio"
--port int (CLI arg) No Port number for HTTP-based transports. Default: 8080

Outputs

Name Type Description
server FastMCP (running) A running MCP server instance accessible via the specified transport protocol

Usage Examples

# Launch MCP server in granular-ops mode via stdio
# dj-mcp granular-ops --transport stdio

# Launch MCP server in recipe-flow mode via SSE on port 8000
# dj-mcp recipe-flow --transport sse --port 8000

# Programmatic usage
from data_juicer.tools.mcp_server import main
# main() parses sys.argv and launches the server

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment