The Development of AI Agents in the Web3 Field: Exploring from Manus to MCP

Recently, a product called Manus, the world's first universal AI Agent, has attracted widespread attention. As an AI tool capable of independent thinking, planning, and executing complex tasks, Manus demonstrates unprecedented versatility and execution power, providing valuable product ideas and design inspiration for the development of AI Agents.

An AI Agent is a computer program that can autonomously make decisions and execute tasks based on the environment, inputs, and predefined goals. Its core components include large language models (LLM), observation and perception mechanisms, reasoning processes, action execution, and memory and retrieval functions. Currently, there are two main development paths for the design patterns of AI Agents: one that emphasizes planning capabilities and another that emphasizes reflective capabilities.

In the Web3 industry, although the development of AI Agents has gone through a period of stagnation, there are still some projects actively exploring. These mainly include three models: the launch platform model represented by Virtuals Protocol, the DAO model represented by ElizaOS, and the commercial company model represented by Swarms. Among them, the launch platform model is currently the most likely to achieve a self-sustaining economic closed loop.

The emergence of Model Context Protocol (MCP) has brought new exploration directions for AI Agents in Web3. On one hand, the MCP Server can be deployed on blockchain networks to solve single point issues and possess censorship resistance; on the other hand, the MCP Server can have the functionality to interact with the blockchain, lowering the technical barrier. In addition, some researchers have proposed a scheme to build an OpenMCP.Network creator incentive network based on Ethereum.

Although the combination of MCP and Web3 theoretically injects decentralized trust mechanisms and economic incentives into AI Agent applications, there are still some challenges with current technology, such as the difficulty of verifying the authenticity of Agent behavior using zero-knowledge proof technology, as well as the efficiency issues of decentralized networks.

The integration of AI and Web3 is an inevitable trend. Although there are still many challenges at present, the industry needs to maintain patience and confidence, continuously exploring the application and development of AI Agents in the Web3 field. In the future, we look forward to seeing a milestone product that can break external doubts and demonstrate the practicality of Web3.