Skip to content

Advanced Network Forensics & Threat Intelligence Platform - Offline PCAP Analysis via Model Context Protocol

License

Notifications You must be signed in to change notification settings

kylecui/NetForensicMCP

Repository files navigation

NetForensicMCP Logo

๐Ÿ” Advanced Network Forensics & Threat Intelligence Platform ๐Ÿ›ก๏ธ

Version License Node.js Wireshark

NetForensicMCP v2.1

๏ผˆFormerly WireMCP, Now Focused on Offline Forensic Analysis๏ผ‰

English | ไธญๆ–‡

NetForensicMCP (formerly WireMCP) is a Model Context Protocol (MCP) server designed to empower Large Language Models (LLMs) with advanced offline network traffic analysis and threat intelligence capabilities. Built on top of Wireshark's tshark, NetForensicMCP provides comprehensive PCAP analysis tools for cybersecurity professionals, threat hunters, and network forensics investigators.

๐Ÿš€ Key Features

Core Analysis Engine

  • Smart Stream Analysis: Intelligent content chunking to handle large PCAP files without token overflow
  • Threat Intelligence Integration: Built-in URLhaus blacklist checking with stream correlation
  • Credential Extraction: Automated detection of plaintext credentials across multiple protocols
  • High-Frequency IP Analysis: Proactive threat hunting through top communicator identification

Advanced Tools

  • get_summary_stats: Protocol hierarchy statistics for traffic composition overview
  • get_conversations: TCP/UDP conversation analysis with stream indexing
  • extract_stream_content: Precise payload extraction with pagination support
  • get_stream_info: Content size estimation to prevent token overflow
  • extract_stream_chunks: Automated large stream segmentation
  • get_top_ips: High-frequency communicator identification for proactive analysis
  • check_threats: Batch IP threat scanning with stream correlation
  • extract_credentials: Multi-protocol credential detection with context
  • capture_packets: Legacy live traffic capture (preserved for compatibility)

๐Ÿ” How It Empowers LLMs

NetForensicMCP transforms complex network forensics into LLM-accessible intelligence by:

  • ๐ŸŽฏ Threat-Driven Analysis: Prioritizes high-risk indicators over raw data processing
  • ๐Ÿ“Š Structured Intelligence: Converts PCAP data into actionable threat intelligence
  • โšก Efficient Investigation: Optimized workflow prevents token exhaustion
  • ๐Ÿ”— Correlation Engine: Links disparate network events into coherent attack narratives
  • ๐Ÿ“ Automated Reporting: Generates comprehensive security reports with IOCs and recommendations

๐Ÿ›ก๏ธ Cybersecurity Use Cases

  • ๐Ÿ•ต๏ธ Threat Hunting: Proactive identification of APT activities and C2 communications
  • ๐Ÿ” Incident Response: Rapid forensic analysis of network evidence
  • ๐Ÿ“‹ Compliance Auditing: Credential leak detection and security gap identification
  • ๐Ÿšจ IOC Extraction: Automated indicator of compromise discovery
  • ๐Ÿ“– Attack Reconstruction: Timeline analysis and attack path visualization

๐Ÿ“‹ Installation

Prerequisites

  • Operating System: Windows, macOS, or Linux
  • Wireshark: Download here (tshark must be in PATH)
  • Node.js: v16+ recommended
  • npm: For dependency management

Setup

  1. Clone the repository:

    git clone https://github.com/kylecui/NetForensicMCP.git
    cd NetForensicMCP
  2. Install dependencies:

    npm install
  3. Launch the MCP server:

    node index.js

Note: NetForensicMCP auto-detects tshark or falls back to common installation paths on all platforms.

โš™๏ธ MCP Client Configuration

Cursor IDE

Edit mcp.json in Cursor โ†’ Settings โ†’ MCP:

{
  "mcpServers": {
    "netforensicmcp": {
      "command": "node",
      "args": [
        "/ABSOLUTE_PATH_TO/NetForensicMCP/index.js"
      ]
    }
  }
}

Claude Desktop

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "wiremcp": {
      "command": "node",
      "args": ["C:\\path\\to\\NetForensicMCP\\index.js"]
    }
  }
}

๐Ÿ”ฌ Example Analysis Workflows

Threat Intelligence Analysis

# Batch threat scanning with stream correlation
check_threats โ†’ extract_credentials โ†’ get_top_ips
โ†“
ip_reputation (parallel) โ†’ ioc_detection โ†’ domain_analysis
โ†“
extract_stream_content (targeted) โ†’ comprehensive_report

Advanced Forensics

# Large PCAP investigation
get_summary_stats โ†’ get_conversations โ†’ get_stream_info
โ†“
extract_stream_chunks โ†’ extract_stream_content (paginated)
โ†“
correlation_analysis โ†’ timeline_reconstruction

๐Ÿ“Š Sample Output

Threat Analysis Report

โš ๏ธ  THREATS DETECTED (2):
๐Ÿšจ 192.168.1.100 - Streams: [tcp:0, tcp:2, udp:1]
๐Ÿšจ 10.0.0.50 - Streams: [tcp:5]

๐Ÿ“‹ RECOMMENDED NEXT STEPS:
1. Use threat intelligence tools to analyze these IPs
2. Extract stream content for streams containing these IPs  
3. Focus investigation on: 192.168.1.100, 10.0.0.50

Stream Content Analysis

Content of tcp stream 0 (chars 0-15000 of 45230):
POST /api/upload HTTP/1.1
Host: suspicious-domain.com
Authorization: Bearer eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9...

[TRUNCATED - More content available. Use offset=15000 to get the next chunk.]

๐ŸŽฏ Advanced Features

Smart Token Management

  • Intelligent Chunking: Automatic content segmentation prevents API limits
  • Pagination Support: Seamless navigation through large datasets
  • Size Estimation: Proactive content size assessment
  • Parallel Processing: Efficient batch operations

Threat Intelligence Integration

  • URLhaus Integration: Comprehensive malware URL database checking
  • Stream Correlation: Links threats to specific communication flows
  • IOC Extraction: Automated indicator discovery and validation
  • Proactive Scanning: Top communicator threat assessment

๐Ÿ› ๏ธ Architecture

NetForensicMCP v2.1 implements an optimized investigation workflow:

  1. ๐Ÿ“ก Reconnaissance Phase: Low-token traffic overview
  2. ๐Ÿ” Batch Scanning Phase: Parallel threat detection
  3. ๐Ÿง  Intelligence Phase: Deep threat correlation
  4. ๐Ÿ“‹ Planning Phase: Strategic analysis targeting
  5. ๐ŸŽฏ Payload Phase: Precision content extraction
  6. ๐Ÿ“Š Reporting Phase: Comprehensive findings synthesis

๐Ÿš€ Roadmap

  • ๐Ÿ”Œ Extended IOC Sources: Integration with VirusTotal, AlienVault OTX
  • ๐Ÿค– ML-Powered Analysis: Behavioral pattern recognition
  • ๐Ÿ“ˆ Timeline Visualization: Interactive attack reconstruction
  • ๐Ÿ”„ Enhanced Automation: Advanced workflow automation capabilities
  • ๐Ÿ“ฑ Web Dashboard: Browser-based analysis interface

๐Ÿค Contributing

We welcome contributions! Please see our contribution guidelines for details.

Areas for Contribution:

  • Threat Intelligence Sources: Additional IOC providers
  • Protocol Analyzers: New credential extraction methods
  • Performance Optimization: Large PCAP handling improvements
  • Documentation: Use cases and tutorials

๐Ÿ“‹ Documentation

๐Ÿ“„ License

This project is licensed under the MIT License - see the LICENSE file for details.

๐Ÿ“‹ Changelog

See CHANGELOG.md for detailed version history and release notes.

Original Attribution

Based on the original WireMCP project by 0xkoda with significant enhancements for offline analysis and threat intelligence integration. We extend our gratitude to the original author for providing the foundational MCP framework and live capture capabilities that made this advanced forensics platform possible.

๐Ÿ™ Acknowledgments

  • 0xkoda: Original WireMCP creator - thank you for the foundational live capture framework
  • Wireshark Team: For the excellent tshark packet analysis engine
  • Model Context Protocol Community: For the MCP framework and specifications
  • URLhaus (abuse.ch): For providing comprehensive threat intelligence data
  • Cybersecurity Community: For continuous feedback and improvement suggestions

โšก Ready to revolutionize your network forensics? Get started with NetForensicMCP v2.1 today!

About

Advanced Network Forensics & Threat Intelligence Platform - Offline PCAP Analysis via Model Context Protocol

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •