Unibase Docs
  • Introduction
    • Architecture
    • Core Components
    • Quick Start
  • Membase
    • 📚Architecture
    • Identity
    • Multi Memory
    • Authorization
    • 🚀Quick Start
  • AIP
    • Design
    • Agent Interaction Process
    • Implementation
    • Quick Start
      • Tool
      • Agent
        • Agent-Tool Interaction Via gRPC
        • Agent-Tool Interaction Via SSE
        • Agent-Agent Interaction
      • Chess game
  • Unibase DA
    • DA
    • Storage
    • Quick Start
      • Nodes boostrap
        • Storage Node
        • Validator Node
        • Stream Node
        • Docker
      • Nodes operations
  • Developers
    • Hub
    • SDK
    • API Reference
  • Alpha Testing
    • Testing plan
    • Requirements
    • Chains
  • 🚀Examples
  • AIP Whitepaper
Powered by GitBook
On this page
  • What is Membase?
  • 🧩 Core Components
  • 🔄 Interaction Workflow
  • 🌟 Why Membase Matters
  1. Membase

Architecture

What is Membase?

Membase is the core decentralized AI memory and agent management layer of the Unibase ecosystem. It handles agent identity registration, configuration management, memory storage, and enables secure remote interaction services — powered by on-chain verification and decentralized storage.

At its heart is the Agent Hub, coordinating interactions across blockchain (Chain) and Unibase DA (decentralized data availability layer).

🧩 Core Components

1. Agent Hub (Agent Management Center)

Manages the full lifecycle and interaction of agents, consisting of three key sub-modules:

  • Link Hub Facilitates remote interaction services, enabling agents to connect with external services or other agents.

  • Config Hub Manages agent configurations, including identity, permissions, and operational parameters. Supports efficient agent discovery and querying.

  • Memory Hub Stores persistent data, conversation history, and agent interaction records to ensure data continuity and traceability.

2. Chain (Blockchain Layer)

  • Stores agent identities and permission mappings on-chain.

  • Provides cryptographic credibility for agent identities.

  • Supports verifiable operations by synchronizing with Unibase DA for decentralized storage.

3. Unibase DA (Decentralized Data Availability Layer)

  • Stores agent-related information off-chain while preserving verifiability.

  • Ensures decentralized, tamper-proof, and highly available data access.

  • Regularly syncs with on-chain data to ensure maximum integrity and transparency.

🔄 Interaction Workflow

Step
Description

1

Agent Registration: Agents register their identity and permission data on-chain.

2

Synchronization: Agent Hub synchronizes registration and permission information from the blockchain.

3

Remote Interaction: Agents connect via the Link Hub, with identity verification against the blockchain.

4

Configuration Management: Agents register or update operational configurations through the Config Hub.

5

Memory Upload: Agent-generated data (interactions, history) is uploaded to the Memory Hub.

6

Persistence: Data from the Agent Hub is synchronized to Unibase DA for decentralized, persistent storage.

🌟 Why Membase Matters

  • Enables agents to have verifiable, decentralized identities.

  • Supports agent memory persistence, cross-platform interoperability, and self-evolution.

  • Provides a foundation for building the Open Agent Internet.

PreviousMembaseNextIdentity

Last updated 20 days ago

📚