About
The EVM MCP Server offers a unified Model Context Protocol interface that lets AI agents read and write across 30+ EVM-compatible networks. It supports balances, token transfers, contract calls, ENS resolution, and comprehensive transaction management.
Capabilities
EVM MCP Server
The EVM MCP Server is a fully‑featured Model Context Protocol (MCP) backend that unlocks blockchain functionality for AI assistants. By exposing a unified set of tools and resources, it lets agents query state, read contracts, transfer tokens, and perform smart‑contract interactions across more than thirty EVM‑compatible networks—including Ethereum mainnet, Optimism, Arbitrum, Base, Polygon, zkSync Era, and many others. This breadth of coverage removes the need for separate RPC endpoints or custom adapters; a single MCP call can reach any supported chain, dramatically simplifying AI‑driven blockchain workflows.
What makes this server especially valuable is its consistent interface. Every tool accepts the same argument shapes, and address parameters automatically resolve ENS names (e.g., ) to raw addresses. This feature reduces friction for developers and users alike, allowing natural language instructions such as “transfer 10 USDC to vitalik.eth” to be translated into precise blockchain calls. The server also bundles common token operations—ERC20 transfers, ERC721 ownership checks, ERC1155 batch moves—and provides rich metadata queries (token name, symbol, decimals, supply) and balance checks for native coins. By offering both read‑only and signed write services (via private key signing), the MCP server covers the full spectrum of on‑chain interactions that an AI assistant might need.
Key capabilities include:
- Multi‑chain data access: read blocks, transactions, receipts, and chain info (block number, chain ID) for any supported network.
- Token services: ERC20 balances and transfers, ERC721 ownership verification and transfer, ERC1155 batch handling.
- Smart contract interaction: view/pure calls, transaction signing, event log filtering, and contract verification to distinguish EOAs from contracts.
- Transaction utilities: gas estimation, native token transfers, status checks, and detailed error reporting.
- ENS resolution: all address inputs support human‑readable names, automatically mapped to addresses behind the scenes.
In practice, this MCP server enables a variety of real‑world scenarios. A financial AI could automatically fetch portfolio balances across chains, reconcile token holdings, and execute trades on behalf of users. A game bot might read NFT ownership data from Polygon or zkSync and trigger in‑game rewards. A compliance tool could audit transaction histories across multiple networks for regulatory reporting, all without custom RPC handling or chain‑specific SDKs. Developers can embed the server into existing AI workflows, allowing agents to discover and invoke blockchain services as first‑class tools in their reasoning pipelines.
Because the server is written in TypeScript and leverages the Viem library, it benefits from strong type safety and up‑to‑date EVM tooling. The design emphasizes simplicity, extensibility, and security: each tool is self‑contained, can be extended with new networks or token standards, and uses secure signing practices for write operations. This combination of a rich feature set, cross‑chain reach, and developer‑friendly interface makes the EVM MCP Server a standout solution for anyone looking to integrate blockchain capabilities into AI assistants.
Related Servers
MarkItDown MCP Server
Convert documents to Markdown for LLMs quickly and accurately
Context7 MCP
Real‑time, version‑specific code docs for LLMs
Playwright MCP
Browser automation via structured accessibility trees
BlenderMCP
Claude AI meets Blender for instant 3D creation
Pydantic AI
Build GenAI agents with Pydantic validation and observability
Chrome DevTools MCP
AI-powered Chrome automation and debugging
Weekly Views
Server Health
Information
Tags
Explore More Servers
Screenshot Website Fast MCP Server
Fast, AI‑optimized web page screenshots in 1072x1072 tiles
WiseVision ROS2 MCP Server
Easily connect AI to ROS2 with zero‑friction stdio MCP
FastMCP Example Server
Run your MCP server with FastMCP and integrate it into Claude Desktop
MCP SSE Servers
Reference implementations for Model Context Protocol over Server‑Sent Events
MediaWiki MCP Server
Seamless Wikipedia API integration for LLMs
Wikipedia MCP Image Crawler
Search and retrieve public domain images from Wikipedia Commons