Langgraph MCP Agents - MCP Servers

██╗      █████╗ ███╗   ██╗ ██████╗  ██████╗ ██████╗  █████╗ ██████╗ ██╗  ██╗
██║     ██╔══██╗████╗  ██║██╔════╝ ██╔════╝ ██╔══██╗██╔══██╗██╔══██╗██║  ██║
██║     ███████║██╔██╗ ██║██║  ███╗██║  ███╗██████╔╝███████║██████╔╝███████║
██║     ██╔══██║██║╚██╗██║██║   ██║██║   ██║██╔══██╗██╔══██║██╔═══╝ ██╔══██║
███████╗██║  ██║██║ ╚████║╚██████╔╝╚██████╔╝██║  ██║██║  ██║██║     ██║  ██║
╚══════╝╚═╝  ╚═╝╚═╝  ╚═══╝ ╚═════╝  ╚═════╝ ╚═╝  ╚═╝╚═╝  ╚═╝╚═╝     ╚═╝  ╚═╝

███╗   ███╗ ██████╗██████╗      █████╗  ██████╗ ███████╗███╗   ██╗████████╗
████╗ ████║██╔════╝██╔══██╗    ██╔══██╗██╔════╝ ██╔════╝████╗  ██║╚══██╔══╝
██╔████╔██║██║     ██████╔╝    ███████║██║  ███╗█████╗  ██╔██╗ ██║   ██║
██║╚██╔╝██║██║     ██╔═══╝     ██╔══██║██║   ██║██╔══╝  ██║╚██╗██║   ██║
██║ ╚═╝ ██║╚██████╗██║         ██║  ██║╚██████╔╝███████╗██║ ╚████║   ██║
╚═╝     ╚═╝ ╚═════╝╚═╝         ╚═╝  ╚═╝ ╚═════╝ ╚══════╝╚═╝  ╚═══╝   ╚═╝

🤖 Autonomous Security Agents with LangGraph & MCP

Multi-agent system for security operations, code analysis, and Mantis defense integration

🏗️ Architecture

graph TD
    A[User Input] --> B[LangGraph Orchestrator]
    B --> C[Code Analysis Agent]
    B --> D[Knowledge Agent]
    B --> E[Mantis Defense Agent]
    B --> F[Red Team Agent]
    B --> G[Research Agent]

    C --> H[MCP Tool Layer]
    D --> H
    E --> H
    F --> H
    G --> H

    H --> I[Mantis MCP Server]
    H --> J[Memory MCP Server]
    H --> K[Learning Loop MCP]

    I --> L[Security Operations]
    J --> L
    K --> L

    style B fill:#f9f,stroke:#333,stroke-width:4px
    style H fill:#bbf,stroke:#333,stroke-width:2px
    style L fill:#bfb,stroke:#333,stroke-width:2px

Multi-Agent System with LangGraph orchestration, MCP tool integration, and persistent state management.

🌟 Features

Specialized Security Agents

🔍 Code Analysis Agent - Static analysis, vulnerability detection, security patterns
🧠 Knowledge Agent - Memory persistence, context retrieval, pattern matching
🛡️ Mantis Integration Agent - Defensive prompt injection, LLM attack detection
⚔️ Red Team Agent - Automated penetration testing, attack simulation
🔬 Research Agent - Threat intelligence, CVE analysis, security research
🔄 Workflow Agent - Multi-agent orchestration, task coordination

MCP Integration

Model Context Protocol - Unified tool interface for Claude and other LLMs
Real-time Communication - WebSocket and HTTP streaming support
State Management - Persistent conversation context across sessions
Tool Chaining - Multi-step automated workflows

LangGraph Capabilities

Visual Debugging - LangGraph Studio integration for graph visualization
State Checkpointing - Pause, resume, and replay agent execution
Human-in-the-Loop - Interactive approval gates for sensitive operations
Parallel Execution - Concurrent agent processing for performance

Getting Started

Install dependencies, along with the LangGraph CLI, which will be used to run the server.

cd path/to/your/app
pip install -e . "langgraph-cli[inmem]"

(Optional) Customize the code and project as needed. Create a .env file if you need to use secrets.

cp .env.example .env

If you want to enable LangSmith tracing, add your LangSmith API key to the .env file.

# .env
LANGSMITH_API_KEY=lsv2...

Start the LangGraph Server.

langgraph dev

For more information on getting started with LangGraph Server, see here.

How to customize

Define runtime context: Modify the Context class in the graph.py file to expose the arguments you want to configure per assistant. For example, in a chatbot application you may want to define a dynamic system prompt or LLM to use. For more information on runtime context in LangGraph, see here.
Extend the graph: The core logic of the application is defined in graph.py. You can modify this file to add new nodes, edges, or change the flow of information.

Development

While iterating on your graph in LangGraph Studio, you can edit past state and rerun your app from previous states to debug specific nodes. Local changes will be automatically applied via hot reload.

Follow-up requests extend the same thread. You can create an entirely new thread, clearing previous history, using the + button in the top right.

For more advanced features and examples, refer to the LangGraph documentation. These resources can help you adapt this template for your specific use case and build more sophisticated conversational agents.

LangGraph Studio also integrates with LangSmith for more in-depth tracing and collaboration with teammates, allowing you to analyze and optimize your chatbot's performance.

MCP Servers