Aureus AI
  • Executive Summary
  • INTRODUCTION
    • Background
    • Vision & Mission
  • PROBLEM STATEMENT
    • Challenges in DeFi Automation
    • Fragmented Agent Marketplaces
  • Solution Overview
    • Aureus AI Platform
    • Key Features
  • Aureus AI Platform Details
    • Custom AI Agents
    • AI Chat Agent
    • AI Agent Certification & Validation
    • One-Stop AI Agent Marketplace
    • Instant AI Agent Creation & No-Code Customization
    • MCP Server Overview
  • Tokenomics
    • Aureus Token ($AUREUS) Utility
    • Token Distribution & Allocation
    • Revenue Model Integration with Tokenomics
    • Long-Term Vision for $AUREUS
  • Technology Architecture
    • Blockchain Integration
    • AI and Machine Learning
  • Business Model
    • Revenue Streams
    • Subscription Model for AI Agent Creation
    • AI Agent Marketplace & Revenue Sharing
    • Partnerships & Collaborations
    • Summary
    • Final Thoughts
  • Roadmap
    • Phase 1
    • Phase 2
    • Phase 3
    • Phase 4
    • Phase 5
    • Summary
    • Final Thoughts
  • Appendices
    • Glossary of Terms
    • Technical Specifications
    • Legal Considerations
    • Final Thoughts
  • Socials
    • Website
    • X (Formerly Twitter)
    • Telegram
Powered by GitBook
On this page
  1. Aureus AI Platform Details

Custom AI Agents

Overview

Custom AI Agents are the core utility of Aureus AI — allowing users to create, configure, and launch autonomous trading agents based on personalized strategies. These agents act on behalf of the user, executing buy, sell, or portfolio management decisions in real-time, using secure DeFi protocols.

Each agent is isolated in its own custodial wallet, operates under rule-based logic, and is driven by a lightweight AI engine that interprets user-defined parameters to make execution decisions.


Key Components

  • Wallet-Scoped Deployment: Each AI agent is deployed with a dedicated custodial wallet, allowing isolated trade execution and individualized fund control.

  • User-Defined Parameters: Users define the agent’s behavior via an intuitive interface, including:

    • Risk Tolerance (Low, Medium, High)

    • Trading Strategy (Scalping or Swing)

    • Investment Mode (Predefined or User-Defined)

    • DCA Preferences (enabled/disabled, repetition count)

    • Take-Profit / Stop-Loss percentages

    • Auto-exit Conditions

  • Autonomous Logic Execution: Once activated, agents monitor market conditions through integrated APIs and execute decisions based on the configured parameters.

  • Tool Integration via MCP: Agent actions like “check token price,” “assess liquidity,” or “initiate trade” are routed through secure, schema-validated tools registered on the Model Context Protocol.


How It Works (Flow)

  1. User creates an agent and sets preferences.

  2. A custodial wallet is deployed for the agent.

  3. The agent interprets user inputs and continuously monitors the market.

  4. When a predefined condition is met (e.g., price drop of 10%), the agent calls appropriate MCP tools like getTokenPrice → walletExecute.

  5. Trades are executed securely on-chain.


Access Control

  • Free Tier: 1 Agent with limited parameters

  • Higher Tiers: Unlock more agents, advanced condition combinations, and faster execution logic

PreviousKey FeaturesNextAI Chat Agent

Last updated 6 days ago