threat-hunting
A minimal, modular MCP server that equips your AI with practical capabilities for real-world threat hunting workflows.
claude mcp add --transport stdio thorcollective-threat-hunting-mcp-server python -m threat_hunting_server \ --env THREAT_HUNTING_CONFIG="Path to configuration file or YAML/JSON content" \ --env THREAT_HUNTING_LOG_LEVEL="INFO|DEBUG|WARN (default: INFO)"
How to use
This MCP server focuses on behavioral threat hunting, prioritizing TTP-based detections over static indicators. It leverages a behavioral hunting framework to analyze host and network activity for patterns that indicate adversary techniques. Use the server to run hypothesis-driven hunts, consult the built-in behavior libraries, and integrate MITRE ATT&CK mappings to identify gaps in detections. The tooling supports querying for behavioral patterns, evaluating hunter hypotheses, and generating reports that emphasize durable, behavior-based detections.
How to install
Prerequisites:
- Python 3.8+ installed on the host
- Access to install Python packages (pip)
- A configuration file or environment variables to configure hunting behavior
Installation steps:
-
Create a virtual environment (optional but recommended): python -m venv venv source venv/bin/activate # Unix/macOS venv\Scripts\activate # Windows
-
Install required Python package(s) (replace with actual package if different): pip install threat-hunting-server # placeholder package name
-
Prepare configuration:
- Create a config file (config.yaml or config.json) with your environment specifics
- Or set environment variables (THREAT_HUNTING_CONFIG, THREAT_HUNTING_LOG_LEVEL)
-
Run the MCP server: python -m threat_hunting_server
-
Verify startup logs indicate the server is listening and ready to accept MCP requests.
Additional notes
Tips and common considerations:
- Use behavior-centric rules and MITRE ATT&CK mappings to align hunts with tactics and techniques (e.g., T1003 LSASS memory access, T1021 lateral movement).
- Configure THREAT_HUNTING_CONFIG to point to your detection library and heuristics.
- Enable verbose logging (THREAT_HUNTING_LOG_LEVEL=DEBUG) while developing hunts, then switch to INFO or WARN in production to reduce log noise.
- If integrating with SIEMs or external telemetry, ensure authentication and encryption are properly configured (e.g., TLS, API keys).
- Regularly refresh threat-hunting knowledge bases (HEARTH integration, hypothesis libraries) to keep detections up to date.
Related MCP Servers
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
mcp-yfinance
Real-time stock API with Python, MCP server example, yfinance stock analysis dashboard
pentestMCP
pentestMCP: AI-Powered Penetration Testing via MCP, an MCP designed for penetration testers.
mcp-panther
Write detections, investigate alerts, and query logs from your favorite AI agents
fastmcp-threatintel
AI-Powered Threat Intelligence MCP tool
opencti_mcp_server
Natural language interface to OpenCTI threat intelligence. Built with Claude Code for $22. Part of Cooper Cyber Coffee.