Phalcon MCP Server

This MCP server integrates with the BlockSec platform to provide blockchain transaction analysis tools via the Model Context Protocol (MCP).

The Model Context Protocol (MCP) is a protocol for AI model integration, allowing AI models to access external tools and data sources.

Components

Tools

Transaction Analysis

Trace
- Trace the different calls of a transaction on a blockchain along with gas usage metrics
- Parameters: chainId (required), transactionHash (required)
Profile
- Profile a transaction on a blockchain with details about the transaction, flow of funds and token information
- Parameters: chainId (required), transactionHash (required)
AddressLabel
- Get human readable labels for contract addresses like tokens, protocols, and other on-chain entities
- Parameters: chainId (required), transactionHash (required)
BalanceChange
- Retrieve detailed balance change information for a transaction
- Parameters: chainId (required), transactionHash (required)
StateChange
- Retrieve detailed information about state changes like storage variables in contracts for a transaction
- Parameters: chainId (required), transactionHash (required)
TransactionOverview
- Comprehensive overview of a transaction by aggregating data from all available analysis tools
- Parameters: chainId (required), transactionHash (required)

Chain Information

GetChainIdByName
- Get the chain ID for a blockchain by name, chain, or chainSlug
- Parameters: name (required)

Getting Started

Installation

Using Go Install

go install github.com/mark3labs/phalcon-mcp@latest

Usage

Start the MCP server:

phalcon-mcp serve

Check the version:

phalcon-mcp version

Using as a Package

You can import the server in your Go projects:

Stdio Mode

import "github.com/mark3labs/phalcon-mcp/server"

func main() {
    // Create a new server with version
    s := server.NewServer("1.0.0")
    
    // Start the server in stdio mode
    if err := s.ServeStdio(); err != nil {
        log.Fatalf("Server error: %v", err)
    }
}

In-Process Mode

For in-process usage with the mcp-go client library:

import (
    "context"
    "log"
    
    "github.com/mark3labs/mcp-go/client"
    "github.com/mark3labs/mcp-go/client/transport"
    "github.com/mark3labs/phalcon-mcp/server"
)

func main() {
    // Create the Phalcon MCP server
    phalconServer := server.NewServer("1.0.0")

    // Create an in-process transport using the server's MCPServer
    inProcessTransport := transport.NewInProcessTransport(phalconServer.GetMCPServer())

    // Create an MCP client using the in-process transport
    mcpClient := client.NewMCPClient(inProcessTransport)

    // Start the transport
    ctx := context.Background()
    if err := mcpClient.Connect(ctx); err != nil {
        log.Fatalf("Failed to connect: %v", err)
    }
    defer mcpClient.Close()

    // Initialize the client
    if err := mcpClient.Initialize(ctx); err != nil {
        log.Fatalf("Failed to initialize: %v", err)
    }

    // List available tools
    tools, err := mcpClient.ListTools(ctx)
    if err != nil {
        log.Fatalf("Failed to list tools: %v", err)
    }

    // Use the tools...
    result, err := mcpClient.CallTool(ctx, "get-chain-id-by-name", map[string]any{
        "name": "ethereum",
    })
    if err != nil {
        log.Fatalf("Failed to call tool: %v", err)
    }
}

Usage with Model Context Protocol

To integrate this server with apps that support MCP:

{
  "mcpServers": {
    "phalcon": {
      "command": "phalcon-mcp",
      "args": ["serve"]
    }
  }
}

Docker

Running with Docker

You can run the Phalcon MCP server using Docker:

docker run -i --rm ghcr.io/mark3labs/phalcon-mcp:latest serve

Docker Configuration with MCP

To integrate the Docker image with apps that support MCP:

{
  "mcpServers": {
    "phalcon": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/mark3labs/phalcon-mcp:latest",
        "serve"
      ]
    }
  }
}

License

MIT

phalcon-mcp