Alchemy MCP Server
Get started with the Alchemy Model Context Protocol (MCP) server.
The Model Context Protocol (MCP) is an open protocol designed to allow AI agents, IDEs, and automation tools to consume, query, and analyze structured data through context-aware APIs.
This server exposes Alchemy’s APIs through the MCP protocol, allowing AI agents and tools to directly query and analyze blockchain data.
Quick Setup
To quickly set up the Alchemy MCP server, add this configuration to your MCP config file (typically found in Claude Desktop or Cursor settings):
The Alchemy MCP server currently only supports STDIO transport for local MCP client use. Remote connection to the MCP server via Streamable-Http transport is currently in development.
Want to learn more about Alchemy’s MCP Server? Visit the official repo here or read more below!
Key Features of Alchemy MCP Server
- Query token prices and price history (including flexible time frame queries)
- Get NFT ownership information and contract data
- View transaction history across multiple networks
- Check token balances across multiple blockchain networks
- Retrieve detailed asset transfers with filtering
How You Would Use The Alchemy MCP Server
AI agents like Claude, Cursor, and ChatGPT are “chat agents” by default. To work with MCP, they act as MCP clients.
Within an MCP client like Cursor, you can define an mcp.json
configuration file that specifies:
- Which MCP servers the agent should connect to
- What environment variables and arguments to pass so the server runs correctly
Once connected, the client instantly gains access to all the tools defined in the server, enabling the LLM to use them in real time.
The main focus right now for Alchemy MCP is on:
- Local developers using Claude Code or Cursor
- Developers running AI agents locally that need access to blockchain data