GhostWall MCP Server
Enables offline multi-agent security orchestration by providing tools for network traffic analysis, log correlation, threat scoring, and firewall mitigation via JSON-RPC.
README
๐ก๏ธ GhostWall - AI-Based Cyber Threat Detection Framework
GhostWall is a full-stack, completely offline, AI-powered Cyber Threat Detection Framework. Designed with a premium, high-visibility security dashboard, GhostWall serves as a mock defensive operations console. It demonstrates how multi-agent security orchestrations (ADK) can scan network traffic, correlate audit logs, evaluate threat payloads, and trigger active firewall blocksโentirely locally and without external API dependencies.
๐ Live Demo URL: https://ghostwall.vercel.app (Build ready for Vercel deployment)
๐ Visual Walkthrough & Screenshots
๐ป System Security HUD
The main dashboard features glassmorphic control cards, a system health score index, real-time restricted IP counters, and throughput meters.

๐น Real-Time Agentic Response Simulation
Watch the local multi-agent system respond step-by-step to a volumetric DDoS attack and sanitize inputs in the payload scanner:

๐ฏ Key Features
- ADK Multi-Agent System: Orchestrates specialized security agents working in a collaborative pipeline.
- Model Context Protocol (MCP): Implements an offline JSON-RPC MCP structure to register resources and tools that agents can query and trigger.
- Threat Simulator: Interactively trigger simulated attack vectors (DDoS, Brute Force, SQL Injection, XSS, Port Scans) and witness immediate real-time defensive reactions.
- Payload Sanitizer Sandbox: An interactive playground demonstrating strict static sanitization methods (SQLi, XSS, Traversal, Command Injection) with zero risk of code execution.
- CLI Utility Skills: Built-in script tools for directory log scanning and independent payload validation.
- Modern Obsidian Styling: Sleek cyberpunk theme featuring glowing neon status indicators, interactive SVG icons, and a custom command-line log viewer.
๐ ๏ธ Tech Stack
- Frontend: React 18, Vite 5, Lucide Icons, Vanilla HSL CSS Variables (Obsidian glassmorphism design system)
- Backend: Node.js, Express, Body Parser, CORS
- Type Safety: TypeScript 5
- Deployment: Vercel (integrated with Serverless Functions for full-stack compatibility)
- Orchestration: ADK (Agent Development Kit) simulation engine
๐๏ธ System Architecture
GhostWall models a standard SOC (Security Operations Center) pipeline:
flowchart TD
User([User Simulation Trigger]) -->|Select DDoS/SQLi/Brute Force| App[GhostWall Dashboard Client]
App -->|POST /api/simulation/run| Express[Express.js Serverless Backend]
subgraph Multi-Agent System [ADK Orchestration Engine]
Express --> Manager[AgentManager]
Manager -->|Step 1| NetAgent[Network Traffic Analyzer Agent]
Manager -->|Step 2| LogAgent[Log Parser & Correlation Agent]
Manager -->|Step 3| ThreatAgent[Threat Scoring Agent]
Manager -->|Step 4| MitAgent[Response Mitigation Agent]
end
subgraph Local MCP Server [Model Context Protocol Server]
NetAgent -->|Call Tool| Tool1[network_traffic_analyzer]
LogAgent -->|Call Tool| Tool2[log_correlator]
ThreatAgent -->|Call Tool| Tool3[payload_sanitize_validator]
MitAgent -->|Call Tool| Tool4[firewall_mitigator]
Manager -->|Read Resource| Resource1[ghostwall://threats/logs]
Manager -->|Read Resource| Resource2[ghostwall://system/status]
end
Tool4 -->|Enforce Rule| Blacklist[(Firewall Blacklist / Memory Logs)]
Blacklist -->|Report Status| App
๐ค Agent Directory & Roles
- Network Traffic Analyzer Agent: Scans connections for volumetric patterns (e.g. DDoS volume threshold anomalies).
- Log Parser & Correlation Agent: Links authentication failures with scanning events to spot multi-stage campaigns.
- Threat Classification & Scoring Agent: Inspects exploit payloads and assigns risk index scores (0-100).
- Automated Response Mitigation Agent: Determines containment actions (
BLOCK_IP,RATE_LIMIT,ISOLATE_SUBNET).
๐ Project Directory Structure
โโโ api/
โ โโโ index.ts # Vercel serverless entry point
โโโ assets/
โ โโโ initial_dashboard_view.png
โ โโโ ghostwall_threat_simulation_demo.webp
โโโ server/
โ โโโ index.ts # Express application & API routing
โ โโโ mcp-server.ts # Mock MCP JSON-RPC Server
โ โโโ adk/
โ โ โโโ agent-manager.ts # Multi-agent orchestrator loop
โ โ โโโ network-agent.ts # Network security agent
โ โ โโโ log-agent.ts # Correlation agent
โ โ โโโ threat-agent.ts # Scoring agent
โ โ โโโ mitigation-agent.ts # Firewall mitigation agent
โ โโโ scripts/
โ โโโ log-collector.ts # CLI Log Analyzer skill script
โ โโโ payload-validator.ts # Standalone static payload scanner
โโโ src/
โ โโโ App.tsx # Dashboard interface component
โ โโโ main.tsx # React DOM bootstrapper
โ โโโ index.css # Obsidian cyber theme CSS styles
โ โโโ vite-env.d.ts # Vite client type bindings
โโโ index.html # HTML entry (Outfit & JetBrains Fonts)
โโโ package.json # Project dependencies & build automation
โโโ tsconfig.json # TypeScript compilation config
โโโ vercel.json # Vercel deployment rewrite routes
โโโ vite.config.ts # Vite configuration with proxy settings
๐ Quick Setup & Installation
Prerequisites
- Node.js (v18+)
- npm (v9+)
Local Development
-
Clone the repository and install dependencies:
git clone https://github.com/anshikasingh28072006/GhostWall.git cd GhostWall npm install -
Run the full application locally:
npm run devThis concurrent script launches the React client on http://localhost:3000 and proxy-binds to the Express backend API on http://localhost:3001.
CLI Security Tools
You can execute standalone tools directly in your terminal:
# Run log parser skill
npx ts-node server/scripts/log-collector.ts
# Run payload validator skill
npx ts-node server/scripts/payload-validator.ts "SELECT * FROM users WHERE 1=1"
โก Vercel Deployment Instructions
GhostWall is deployed as a 100% client-side static React application. There is no backend server setup or API key configuration needed:
- Push Changes to GitHub: Ensure all local changes are committed and pushed to your repo.
- Import Project to Vercel:
- Go to Vercel Dashboard and click Add New -> Project.
- Import the
GhostWallrepository.
- Configure Settings:
- Framework Preset: Vite
- Root Directory:
./(leave default) - Build Command:
npm run build(runstsc && vite build) - Output Directory:
dist
- Deploy: Click Deploy. Vercel will build the React bundle and deploy the static site instantly.
๐ฎ Future Improvements
- Add a live interactive threat map visualizing blocked geo-IP markers.
- Implement exportable CSV reports for compliance auditing.
- Expand signature lists to include SQLi-to-shell patterns.
- Integrate WebSockets for live logging feeds.
๐ฅ Author & Contact
- Author: Anshika Singh
- GitHub: @anshikasingh28072006
- Role: Cyber Security Analyst & Developer
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.