mcpField
M
mcp-server-ntopng

mcp-server-ntopng

Local
github.com/marcoegData & Analytics2Updated 84w ago

About mcp-server-ntopng

mcp-server-ntopng lets AI agents query ntopng network-monitoring data, for operators investigating traffic, hosts, and alerts. It uses historical flow and alert data stored in ClickHouse.

Key features

  1. 01Retrieve interface IDs and host locations
  2. 02Rank local and remote talkers
  3. 03Get statistics for alert types
  4. 04Query detailed and top flows
  5. 05Inspect and query the ntopng database

Use cases

  1. 01Review talkers on a selected interface
  2. 02Summarize flow, host, or system alerts
  3. 03Inspect historical flow records
  4. 04List database tables and run queries

MCP support

The MCP server exposes tools for querying ntopng interfaces, hosts, flows, alerts, and database tables to MCP clients. It does not provide resources or prompts.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

mcp-server-ntopng

PyPI - Version

NTOPNG Model Context Protocol Server

README image

A Model Context Protocol server implementation for NTOPNG that enables AI agents to query networks monitoring data using the NTOPNG database.

This MCP Server assumes that ntopng is using ClickHouse to store historical flows and alert. Check ntopng Clickhouse

Tools

  • fetch_ntopng_all_ifids
  • Retrieve all available interface IDs from ntopng.
  • get_ntopng_hosts_location
  • Fetch geographical location and additional info for hosts.
  • fetch_ntopng_top_local_talkers
  • Retrieve the top 10 local talkers for a specified interface.
  • fetch_ntopng_top_remote_talkers
  • Retrieve the top 10 remote talkers for a specified interface.
  • get_ntopng_all_alert_stats
  • Retrieve statistics for all alerts.
  • get_ntopng_flow_alert_stats
  • Retrieve statistics for flow alerts.
  • get_ntopng_host_alert_stats
  • Retrieve statistics for host alerts.
  • get_ntopng_interface_alert_stats
  • Retrieve statistics for interface alerts.
  • get_ntopng_mac_alert_stats
  • Retrieve statistics for MAC alerts.
  • get_ntopng_network_alert_stats
  • Retrieve statistics for network alerts.
  • get_ntopng_snmp_device_alert_list
  • Retrieve a list of SNMP device alerts.
  • get_ntopng_snmp_device_alert_stats
  • Retrieve statistics for SNMP device alerts.
  • get_ntopng_system_alert_stats
  • Retrieve statistics for system alerts.
  • query_ntopng_flows_data
  • Retrieve detailed flows data from the ntopng flows database.
  • get_ntopng_top-k_flows
  • Retrieve top-k flows data from the ntopng flows database.
  • get_ntopng_user_alert_stats
  • Retrieve statistics for user alerts.
  • get_ntopng_flow_devices_stats
  • Retrieve statistics for all flow dev`ices.
  • get_ntopng_sflow_devices_stats
  • Retrieve statistics for all sFlow devices.
  • list_tables_ntopng_database
  • List tables structure of the ntopng database.
  • query_ntopng_database
  • Query the ntopng Clickhouse database.

Status

Works with Claude Desktop app and other MCP compliant hosts and clients.

No support for MCP resources or prompts yet.

Configuration

  1. Create or edit the Claude Desktop configuration file located at:

    • On macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • On Windows: %APPDATA%/Claude/claude_desktop_config.json
  2. Add the following:

{
  "mcpServers": {
    "mcp-ntopng": {
      "command": "/path/to/your/uv-binary",
      "args": ["run", "--with", "mcp-ntopng", "--python", "3.13", "mcp-ntopng"]
      "env": {
        "NTOPNG_HOST": "<ntopng-host>",
        "NTOPNG_DBPORT": "<ntopng-dbport>",
        "NTOPNG_DBUSER": "<ntopng-dbuser>",
        "NTOPNG_DBPASSWORD": "<ntopng-dbpassword>",
        "NTOPNG_SECURE": "true",
        "NTOPNG_VERIFY": "true",
        "NTOPNG_CONNECT_TIMEOUT": "30",
        "NTOPNG_SEND_RECEIVE_TIMEOUT": "300",
        "NTOPNG_API_KEY": "NTOPNG_TOKEN"
      }
    }
  }
}
  1. Replace /path/to/your/uv-binary with the absolute path to the uv executable. Find the path with which uv. This ensures that the correct version of uv is used when starting the server.

  2. Restart Claude Desktop to apply the changes.

Development

  1. Set the environmental variables either in the claude_desktop_config.json file or in a .env file in the root of the repository.
NTOPNG_HOST=localhost
NTOPNG_PORT=9000
NTOPNG_USER=default
NTOPNG_PASSWORD=
  1. Run uv sync to install the dependencies. To install uv follow the instructions here. Then do source .venv/bin/activate.

  2. Install the mcp-ntopng package with uv pip install -e . from the project main directory.

  3. For easy testing, you can run mcp dev mcp_ntopng/mcp_server.py to start the MCP server. CHANGE WITH A PROPER CHAT CLIENT

Environment Variables

The following environment variables are used to configure the database connection:

  • NTOPNG_HOST: The hostname of the ntopng server
  • NTOPNG_DBUSER: The username for Clickhouse DB authentication
  • NTOPNG_DBPASSWORD: The password for Clickhouse DB authentication
  • NTOPNG_API_KEY: The ntopng authentication token.
Optional
  • NTOPNG_DBPORT: The port number of the Clickhouse DB in the ntopng server
    • Default: 9000 if HTTPS is enabled, 8123 if disabled
    • Usually doesn't need to be set unless using a non-standard port
  • NTOPNG_SECURE: Enable/disable a TLS connection
    • Default: false
    • Set to true for a secure TLS connections
  • NTOPNG_VERIFY: Enable/disable SSL certificate verification
    • Default: true
    • Set to `false to disable certificate verification (not recommended for production)
  • NTOPNG_CONNECT_TIMEOUT: Connection timeout in seconds
    • Default: `30
    • Increase this value if you experience connection timeouts
  • NTOPNG_SEND_RECEIVE_TIMEOUT: Send/receive timeout in seconds
    • Default: 300
    • Increase this value for long-running queries

Check TLS Setup in the ntopng documentation for details about setting up a TLS connection to Clickhouse.

Development

Install the package on the local machine:

$ uv sync
$ uv pip install -e .

Run the MCP Inspector

$ cd mcp_ntopng
$ source .env
$ CLIENT_PORT=8077 SERVER_PORT=8078  mcp dev run_mcp_ntopng.py --with clickhouse-driver --with python-dotenv --with uvicorn --with pip-system-certs

Use the local library in Claude Desktop.

Find: /Users/marco/Library/Application\ Support/Claude/claude_desktop_config.json

Edit the claude_desktop_config.json changing the local paths:

{
    "mcpServers": {
      "mcp-ntopng": {
        "command": "/Users/marco/Development/claude/mcp-server-ntopng/.venv/bin/python",
        "args": [
           "/Users/marco/Development/claude/mcp-server-ntopng/run_mcp_ntopng.py"
        ],
        "env": {
          "NTOPNG_HOST": "marcoeg-nod004.ntoplink.com",
          "NTOPNG_DBPORT": "9000",
          "NTOPNG_DBUSER": "default",
          "NTOPNG_DBPASSWORD": "",
          "NTOPNG_SECURE": "false",
          "NTOPNG_VERIFY": "false",
          "NTOPNG_CONNECT_TIMEOUT": "30",
          "NTOPNG_SEND_RECEIVE_TIMEOUT": "300",
          "SELECT_QUERY_TIMEOUT_SECS": "30",
          "NTOPNG_API_KEY": "NTOPNG_TOKEN"
        }
      }
    }
  }

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.