modelscope·@frankausberlin/adminmcp
暂无描述。
A comprehensive MCP (Model Context Protocol) based agent environment for system administration and automation tasks. AdminMCP provides both a CLI tool for managing MCP servers and a server implementation with built-in tools and resources for administrative operations.
AdminMCP is a dual-purpose tool that serves as both a command-line interface (CLI) for managing MCP servers and an MCP server itself. The project aims to create a robust environment for system administration tasks through MCP-compatible agents.
The CLI component (acp) provides easy management of MCP server processes, while the server component offers a collection of tools and resources for common administrative and computational tasks. Future versions will include client functionality for connecting to and interacting with MCP servers.
~/.config/adminmcp/uv package manager (recommended) or pipClone the repository:
git clone https://github.com/yourusername/adminmcp.git
cd adminmcp
Create virtual environment:
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
Install dependencies:
# Using uv (recommended)
uv pip install -e .
# Or using pip
pip install -e .
Verify installation:
acp --help
# Install uv if not already installed
curl -LsSf https://astral.sh/uv/install.sh | sh
# Install AdminMCP
uv pip install -e .
Start the MCP server:
acp server start
# Output: Server started.
Stop the MCP server:
acp server stop
# Output: Server stopped.
Initialize configuration:
acp init
# Output:
# Copied config.adminmcp.json to /home/user/.config/adminmcp/config.adminmcp.json
# Created /home/user/.config/adminmcp/mcp_settings.json
# Configuration initialized.
The easiest way to test the server is using the MCP Inspector:
mcp dev ./src/adminmcp/server/acp_server.py
This starts the MCP Inspector UI where you can interact with all available tools and resources.
The server can be used programmatically through MCP clients. Here are some example tool calls:
# Mathematical operations
result = add(5, 3) # Returns 8
result = multiply(4, 7) # Returns 28
# Resource access
pi_value = get_constant("pi") # Returns 3.141592653589793
is_even = is_even(42) # Returns True
# DateTime information
current = current_datetime()
# Returns: {"datetime": "2024-01-01T12:00:00+00:00", "timezone": "UTC", "unix_timestamp": 1704110400.0}
# Wikipedia lookup
article = wikipedia_article("Python (programming language)")
# Returns article summary JSON
AdminMCP uses a configuration system stored in ~/.config/adminmcp/. The configuration is initialized automatically with the acp init command.
config.adminmcp.jsonTemplate configuration file with default settings:
{
"application": {
"config_folder": "~/.config/adminmcp",
"log_folder": "~/.local/share/adminmcp/logs",
"data_folder": "~/.local/share/adminmcp/data",
"logging_level": "info"
},
"server": {
"port": 8000
},
"client": {}
}
mcp_settings.jsonRuntime configuration for MCP server settings: …