Security Scanner

Provides comprehensive security scanning via Nuclei templates with cluster-wide capabilities and detailed reporting.

python

0

GitHub Stars

python

Language

3 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "marc-shade-security-scanner-mcp": {
      "command": "python",
      "args": [
        "-m",
        "security_scanner.server"
      ],
      "env": {
        "NUCLEI_BIN": "${HOME}/go/bin/nuclei",
        "SCAN_RESULTS_DIR": "${AGENTIC_SYSTEM_PATH:-/opt/agentic}/security-scans"
      }
    }
  }
}

You can run comprehensive security scans across individual targets, networks, and cluster nodes using the Security Scanner MCP Server. It leverages Nuclei templates to detect vulnerabilities and integrates with your Agentic System for centralized management, reporting, and historical analysis.

How to use

Use the Security Scanner MCP Server to perform targeted vulnerability assessments, manage templates, review past findings, and schedule periodic checks. You can scan a single URL or IP, batch scan multiple targets, or scan all nodes in your cluster. Retrieve and review results from any previous scan, and configure templates and severity filters to fit your security needs.

Key usage patterns include initiating a single-target scan, scanning a network or CIDR range, and performing cluster-wide scans to ensure visibility across your entire Agentic deployment. You can also list and update templates, and schedule recurring scans with notifications for new vulnerabilities.

How to install

Prerequisites: Python 3.10+, a Python virtual environment, and access to install Python packages.

Install the Security Scanner MCP locally by navigating to the server directory, creating or activating a virtual environment, and installing in editable mode.

cd ${AGENTIC_SYSTEM_PATH:-/opt/agentic}/mcp-servers/security-scanner-mcp
source ${AGENTIC_SYSTEM_PATH:-/opt/agentic}/.venv/bin/activate
pip install -e .

Configuration and runtime

Configure the MCP client to run the security scanner as a local, stdio-based server. The runtime uses a Python module to start the server and sets environment variables for where to store and locate scan data.

Environment variables you will use include the path to the nuclei binary and the directory where scan results are stored.

Usage overview

To scan a single target, specify the target and optional severity and templates.
To scan multiple targets, provide a list of targets or a CIDR range and adjust concurrency and rate limits.
To scan all cluster nodes, request a cluster-wide scan with a chosen scan type and severity.

Examples

Example commands and calls are shown in the MCP server configuration and tool references. Use the provided commands to perform scans, retrieve results, and manage templates.

Notes on results and reporting

Scan results are stored under the security-scans directory. You can retrieve findings and metadata for each scan, and generate reports in JSON, JSONL, or Markdown formats.

Available tools

scan_target

Scan a single target with Nuclei vulnerability scanner. Supports severity filtering, templates, rate limiting, a timeout, and output formats.

scan_network

Scan multiple targets from a list or CIDR range with configurable concurrency and rate limits.

scan_cluster_nodes

Scan all nodes in the agentic cluster for vulnerabilities using a defined scan type and severity.

list_templates

List available Nuclei templates filtered by tag, severity, or author.

update_templates

Update Nuclei templates to the latest version.

get_scan_results

Retrieve results from a previous scan by scan_id with optional severity filtering.

list_scans

List all past security scans with optional target filtering.

schedule_periodic_scan

Schedule recurring security scans with target lists, interval, severity thresholds, and notifications.