Author: BitMart Research Source: medium
I. Introduction to the concept of MCP
Previously in the field of artificial intelligence, traditional chatbots mostly relied on general dialogue models and lacked personalized role settings, which resulted in their responses often being monotonous and lacking in human touch. To solve this problem, developers introduced the concept of "personality", which is to give AI a specific role, personality, and tone so that its responses are closer to user expectations. However, even if AI has a rich "personality", it is still just a passive responder and cannot actively perform tasks or perform complex operations. Therefore, the open source project Auto-GPT came into being. Auto-GPT allows developers to define a series of tools and functions for AI and register these tools to the system. When a user makes a request, Auto-GPT will generate corresponding operation instructions based on preset rules and tools, automatically perform tasks and return results. This method transforms AI from a passive interlocutor to an active task AI.
Although Auto-GPT has achieved autonomous execution of AI to a certain extent, it still faces problems such as inconsistent tool calling formats and poor cross-platform compatibility. In order to solve these problems, MCP (Model Context Protocol) came into being, which aims to solve the main challenges faced by AI in the development process, especially the complexity of integrating with external tools. The core goal of MCP is to simplify the way AI interacts with external tools, and by providing a unified communication standard, AI can easily call various external services. Traditionally, in order to enable large-scale models to perform complex tasks (such as querying the weather or visiting a web page), developers need to write a lot of code and tool instructions, which greatly increases the difficulty and time cost of development. The MCP protocol significantly simplifies this process by defining standardized interfaces and communication specifications, allowing AI models to interact with external tools more quickly and efficiently.
II. Integration of MCP and AI Agent
MCP and encrypted AI Agent are mutually reinforcing. The difference between the two is that AI Agent mainly focuses on the automated operation of blockchain, smart contract execution and encrypted asset management, emphasizing privacy protection and the integration of decentralized applications. MCP focuses more on simplifying the interaction between AI Agent and external systems, providing standardized protocols and context management, and enhancing cross-platform interoperability and flexibility. Encrypted AI Agent can be more efficiently integrated and operated across platforms through the MCP protocol, thereby improving its execution capabilities.
Previously, AI Agent had certain execution capabilities, such as executing transactions and managing wallets through smart contracts. However, these functions are usually predefined and lack flexibility and adaptability. The core value of MCP is that it provides a unified communication standard for the interaction between AI Agent and external tools (including blockchain data, smart contracts, off-chain services, etc.). This standardization solves the problem of interface fragmentation in traditional development, enables AI Agent to seamlessly connect to multi-chain data and tools, and greatly enhances the autonomous execution capability of AI Agent. For example, DeFi AI Agent can obtain market data in real time and automatically optimize the investment portfolio through MCP. In addition, MCP opens up a new direction for AI Agent, namely the collaboration of multiple AI Agents: through MCP, AI Agents can collaborate according to their functional division of labor, and combine to complete complex tasks such as on-chain data analysis, market forecasting, and risk control management, thereby improving overall efficiency and reliability. On-chain transaction automation: MCP connects various types of transactions and risk control agents in series to solve problems such as slippage, transaction wear and tear, and MEV in transactions, and achieve safer and more efficient on-chain asset management.
III. Related projects
1. DeMCP
DeMCP is a decentralized MCP network. It is committed to providing self-developed open-source MCP services for AI Agents, providing a deployment platform for MCP developers to share commercial benefits, and achieving one-stop access to mainstream large language models (LLMs). Developers can obtain services by supporting stablecoins (USDT, USDC). As of May 8, the market value of its token DMCP is approximately $1.62M.
2.DARK
DARK is an MCP network under a trusted execution environment (TEE) built on Solana. The token $DARK is listed on Binance Alpha, with a market value of approximately US$11.81 million as of May 8. Currently, DARK's first application is under development. It will provide AI Agents with efficient tool integration capabilities through TEE and MCP protocols, allowing developers to quickly access a variety of tools and external services through simple configuration. Although the product has not yet been fully released, users can join the early experience phase by emailing as a candidate, participate in testing and provide feedback.
3. Cookie.fun
Cookie.fun is a platform focused on AIAgent in the Web3 ecosystem, aiming to provide users with a comprehensive AI Agent index and analysis tool. The platform helps users understand and evaluate the performance of different AI Agents by displaying indicators such as AI Agent's mental influence, intelligent following ability, user interaction and on-chain data. On April 24, Cookie.API1.0 was updated and launched a dedicated MCP server, which includes a plug-and-play agent-specific MCP server, designed for developers and non-technical personnel, and does not require any configuration.
4. SkyAI
SkyAI is a Web3 data infrastructure project built on BNB Chain, aiming to build a blockchain-native AI infrastructure by expanding MCP. The platform provides a scalable and interoperable data protocol for Web3-based AI applications. It plans to promote the practical application of AI in blockchain environments by simplifying the development process by integrating multi-chain data access, AI agent deployment, and protocol-level utilities. Currently, SkyAI supports aggregated data sets from BNB Chain and Solana, with a data volume of more than 10 billion rows. In the future, MCP data servers supporting Ethereum mainnet and Base chain will be launched. Its token SkyAI is listed on Binance Alpha, with a market value of approximately US$42.7 million as of May 8.
Fourth, Future Development
As a new narrative of the integration of AI and blockchain, the MCP protocol has shown great potential in improving data interaction efficiency, reducing development costs, enhancing security and privacy protection, and has broad application prospects, especially in scenarios such as decentralized finance. However, most of the current MCP-based projects are still in the proof-of-concept stage and have not yet launched mature products, resulting in a continuous decline in the price of their tokens after they are launched. For example, the price of DeMCP tokens has fallen by 74% in less than a month after they were launched. This phenomenon reflects the market's trust crisis in MCP projects, which is mainly due to the long product development cycle and the lack of practical application. Therefore, how to speed up the development of products, ensure the close connection between tokens and actual products, and improve user experience will be the core issues facing the current MCP projects. In addition, the promotion of the MCP protocol in the crypto ecosystem still faces the challenge of technical integration. Due to the differences in smart contract logic and data structure between different blockchains and DApps, a large amount of development resources are still required to unify and standardize MCP servers.
Despite the above challenges, the MCP protocol itself still shows great market development potential. With the continuous advancement of AI technology and the gradual maturity of the MCP protocol, it is expected to be more widely used in DeFi, DAO and other fields in the future. For example, AI agents can obtain on-chain data in real time through the MCP protocol, perform automated transactions, and improve the efficiency and accuracy of market analysis. In addition, the decentralized nature of the MCP protocol is expected to provide a transparent and traceable operating platform for AI models, promoting the decentralization and assetization of AI assets. As an important auxiliary force in the integration of AI and blockchain, the MCP protocol is expected to become an important engine for promoting the next generation of AI Agents as the technology continues to mature and the application scenarios expand. However, to realize this vision, challenges in technology integration, security, user experience, etc. still need to be addressed.
Jasper
Davin
Jixu
Joy