> For the complete documentation index, see [llms.txt](https://hivex.gitbook.io/hivex/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://hivex.gitbook.io/hivex/under-the-hood/system-overview.md).

# System Overview

Let's take a look at how HiveX works under the hood. Our system combines AI agents, blockchain integration, and a user-friendly interface to create a seamless DeFi experience on MultiversX. This overview will help you understand how all the pieces fit together.

### System Architecture

HiveX is built with a modular architecture that separates concerns while enabling powerful interactions between components:

<figure><img src="/files/rd4r3QMpkJKrVQBxBxdt" alt=""><figcaption></figcaption></figure>

#### Key Components

1. **User Interface**
   * Web application built with Next.js
   * Mobile-responsive design
   * Chat interface for agent interaction
   * Dashboard for portfolio and market data
2. **API Layer**
   * RESTful API endpoints
   * WebSocket connections for real-time updates
   * Authentication and security middleware
   * Request routing and validation
3. **Agent Orchestration**
   * n8n workflows for agent coordination
   * Task delegation and management
   * Response aggregation
   * Error handling and recovery
4. **AI Agents**
   * Specialized agents for different tasks
   * Natural language processing
   * Context management
   * Knowledge retrieval
5. **MultiversX Plugin**
   * Blockchain interaction layer
   * Wallet management
   * Transaction creation and signing
   * Smart contract interaction
6. **Data Providers**
   * Market data integration (CoinGecko, DefiLlama)
   * On-chain data retrieval
   * Historical data analysis
   * Real-time price feeds

### Data Flow

1. **User Input**
   * User sends a request via chat or UI interaction
   * Request is processed and routed to the appropriate agent(s)
2. **Agent Processing**
   * Agent analyzes the request using NLP
   * Relevant information is retrieved from knowledge base
   * Agent determines required actions
3. **Blockchain Interaction**
   * If needed, the agent uses the MultiversX plugin to interact with the blockchain
   * Transactions are created, signed, and broadcast
   * Results are monitored and verified
4. **Response Generation**
   * Agent formulates a response based on action results
   * Response is formatted for user consumption
   * Additional context or suggestions may be added
5. **User Feedback**
   * Response is delivered to the user
   * User can provide feedback or follow-up questions
   * System learns from interactions to improve future responses

### Technical Stack

* **Frontend**: Next.js, React, TypeScript
* **Backend**: Node.js, Express
* **AI**: OpenAI API, Retrieval-Augmented Generation (RAG)
* **Orchestration**: n8n, ElizaOS
* **Blockchain**: MultiversX SDK, xExchange SDK, AshSwap SDK
* **Data**: CoinGecko API, DefiLlama API, MultiversX API

### Security Considerations

* Private keys are never exposed to agents
* All blockchain transactions require explicit user approval
* API keys are securely stored and managed
* Rate limiting and request validation prevent abuse
* Regular security audits and updates

### Extensibility

HiveX is designed to be easily extended:

* New agents can be added to the system
* Additional data providers can be integrated
* The MultiversX plugin can be enhanced with new capabilities
* Custom workflows can be created in n8n

For more detailed information about specific components, check out our [Multi-Agent System](/hivex/multi-agent-system-architecture.md) documentation, [Plugin](/hivex/for-developers/elizaos-plugin-for-hivex.md) details, or [n8n Workflows](/hivex/for-developers/n8n-workflows.md) guide.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://hivex.gitbook.io/hivex/under-the-hood/system-overview.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.