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.

The OpenScan.AI MCP server wraps the explorer APIs and exposes blockchain data — balances, tokens, NFTs, contract metadata — via MCP so that AI agents and tools (like Claude, Cursor, or other IDEs) can access and analyze it contextually.

Key Features:

• Contextual blockchain data access for AI tools
• Configurable explorer instance URL, so the server can point at any OpenScan.AI instance (e.g. https://xdcscan.io for XDC Network)
• Custom instructions for the MCP host describing how to use the server
• Support for MCP progress notifications for multi-step tool operations, allowing clients to track execution status
• Enhanced user experience: periodic progress updates for long-running API queries (e.g. fetching extensive transaction histories) when requested by the client, improving responsiveness

Use Cases

• Enabling AI agents to analyze blockchain transactions and account balances.
• Automating blockchain data analysis workflows with context-aware APIs.
• Allowing IDEs to fetch contract ABIs and other relevant smart contract information.

Configuration

Run an MCP server locally (via stdio or HTTP/SSE, depending on the implementation you choose) and configure it to point at the OpenScan.AI API base URL:

OPENSCAN_API_URL=https://xdcscan.io/api/v2

Then register the server with your MCP host. Most hosts (Claude Desktop, Cursor, etc.) accept a JSON configuration similar to:

{
  "mcpServers": {
    "openscan": {
      "command": "<command-to-run-your-mcp-server>",
      "env": {
        "OPENSCAN_API_URL": "https://xdcscan.io/api/v2"
      }
    }
  }
}

Once registered, the host’s AI agent can call the server’s tools to look up addresses, transactions, tokens, and contract metadata from XDC Network.

Architecture and Data Flow

sequenceDiagram
    participant AI as MCP Host
    participant MCP as MCP Server
    participant OS as OpenScan.AI Instance
    participant Metadata as Metadata Service

    AI->>MCP: __get_instructions__
    MCP-->>AI: Custom instructions

    AI->>MCP: Tool request (address, tx, token, contract lookup)
    par Concurrent API Calls (when applicable)
        MCP->>OS: Request to OpenScan.AI API
        OS-->>MCP: Primary data response
    and
        MCP->>Metadata: Request to Metadata API (for enriched data)
        Metadata-->>MCP: Secondary data response
    end
    MCP-->>AI: Formatted & combined information