MCP Servers

🚀 00_master_mcp Container Technology Documentation

The 00_master_mcp container serves as the central orchestrator of a multi - agent system, coordinating and managing multiple specialized servers to achieve efficient invocation and integration of system resources and services. It provides a single entry point to reuse all downstream server functions and hierarchical namespace organization tools.

🚀 Quick Start

To start using the 00_master_mcp container, follow these steps:

• Build the image: docker build -t mcp_master .
• Run the container: docker run --network host -d mcp_master
• Access the interface: Access the MCP service via http://<host machine>:8000

✨ Features

1. Centralized Control: Exposes a single API endpoint through FastMCPHost to manage multiple specialized servers uniformly.
2. Namespace Isolation: Uses hierarchical namespaces (e.g., os.linux, cloud.azure) to avoid tool name conflicts and enhance system scalability.
3. Context Aggregation: Coordinates the interaction of specialized servers to enable cross - service function calls.
4. Security Policy: Ensures resource access security through namespace - level permission control.

📦 Installation

The installation process is mainly about building and running the Docker container, which is described in the "Quick Start" section.

💻 Usage Examples

Basic Usage

Here is how to build and run the container:

# Build the image
docker build -t mcp_master .
# Run the container
docker run --network host -d mcp_master

Advanced Usage

You can customize the container according to your needs. For example, modify the Dockerfile, entrypoint.sh, or mcp_host.py files.

📚 Documentation

Container Structure

• Dockerfile: Defines the base image for container building, installs runtime dependencies, and configures the startup script.
• entrypoint.sh: Configures network interface binding and user permission restrictions when the container starts to ensure secure operation.
• mcp_host.py: Implements the specific logic of the MCP protocol, including server registration, request distribution, and response handling.

Namespace Policy

• Top - level domains: Divided by functional categories (e.g., os, cloud, infra, web).
• Service specificity: Uses sub - namespaces under each top - level domain to specify specific technologies (e.g., os.linux, cloud.azure).
• Sub - system granularity: Organizes specific functions through further sub - namespaces to avoid excessive containerized tools.

Hierarchical MCP Example

master (:8000)
 ├─ os
 │  ├─ linux           (01_linux_cli_mcp:8001)
 │  │  ├─ cli.*        # Basic commands and file operations
 │  │  ├─ ceph.*       # Ceph cluster management
 │  │  ├─ nginx.*      # Nginx service configuration
 │  │  ├─ mta.*        # Mail transfer agent operations
 │  │  └─ sftp.*       # SFTP server management
 │  ├─ windows.*     (02_windows_mcp:8002)
 │  └─ macos.*       (10_macos_mcp:8010)
 ├─ cloud
 │  ├─ azure.*       (03_azure_mcp:8003)
 │  └─ gcloud.*      (04_google_cloud_mcp:8004)
 ├─ infra
 │  ├─ vmware.*      (05_vmware_mcp:8005)
 │  └─ k8s.*         (08_k8s_mcp:8008)
 ├─ web
 │  ├─ search.*      (06_web_search_mcp:8006)
 │  └─ browse.*      (07_web_browsing_mcp:8007)
 ├─ workflows
 │  └─ n8n.*         (09_n8n_mcp:8009)
 ├─ trading
 │  └─ freq.*        (11_freqtrade_mcp:8011)
 ├─ cmdb              (12_cmdb_mcp:8012)
 │  ├─ local.*       # Local CSV/SQLite database tools
 │  └─ servicenow.*  # ServiceNow integration tools
 └─ secrets           (13_secrets_mcp:8013)
    ├─ keepass.*     # KeePass backend tools
    ├─ azurekv.*     # Azure Key Vault backend tools
    └─ gcloudsa.*    # Google Cloud Secret Manager backend tools

Code Implementation

Dockerfile

# Base image
FROM python:3.9-slim

# Install runtime dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
    ca-certificates \
 && rm -rf /var/lib/apt/lists/*

# Set the working directory
WORKDIR /app

# Copy files
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

COPY . .

# Configure the startup script
ENTRYPOINT ["./entrypoint.sh"]

entrypoint.sh

#!/bin/sh
set -e

# Bind port 8000 to the host machine
sed -i "s/127.0.0.1/0.0.0.0/g" /app/mcp_host.py

# Run as a non-root user
useradd -m mcp_user
chown -R mcp_user /app
su -c "python3 /app/mcp_host.py" mcp_user

mcp_host.py

import sys
from fast_mcp import FastMCPHost

def main():
    host = FastMCPHost()
    try:
        host.start("0.0.0.0", 8000)
    except KeyboardInterrupt:
        print("\nStopping MCP Host...")
    except Exception as e:
        print(f"Error: {e}")
        sys.exit(1)

if __name__ == "__main__":
    main()