MCP DefiLlama

• Package: @iqai/defillama-mcp
• Purpose: Interacting with DefiLlama, a comprehensive DeFi data aggregator, for protocol analytics and TVL data.

Usage with ADK TypeScript

import {McpDefillama} from "@iqai/adk";

const toolset = McpDefillama()

const tools = await toolset.getTools()

import {McpToolset} from "@iqai/adk";

const toolset = new McpToolset({
  name: "Defillama MCP Client",
  description: "Client for DefiLlama DeFi data aggregator",
  transport: {
    mode: "stdio",
    command: "pnpm",
    args: ["dlx", "@iqai/defillama-mcp"],
    env: {
      PATH: process.env.PATH || "",
    },
  },
})

const tools = await toolset.getTools()

{
  "mcpServers": {
    "defillama-mcp-server": {
      "command": "pnpm",
      "args": ["dlx", "@iqai/defillama-mcp"]
    }
  }
}

Available Tools

Environment Variables

No required environment variables for basic usage.

Optional environment variables:

• DEFILLAMA_API_KEY: Your DefiLlama API key for enhanced access.
• IQ_GATEWAY_URL: Custom IQ Gateway URL for enhanced resolution.
• IQ_GATEWAY_KEY: API key for IQ Gateway access.
• OPENROUTER_API_KEY: API key for OpenRouter LLM integration.
• GOOGLE_GENERATIVE_AI_API_KEY: Google generativeAI API key (alternative to OpenRouter).
• LLM_MODEL: Model name for LLM integration (default: openai/gpt-4.1-mini).

Usage Examples

Here's a complete example of using MCP DefiLlama with an ADK agent:

import { McpDefillama, AgentBuilder } from "@iqai/adk";
import dotenv from "dotenv";

dotenv.config();

async function main() {
  // Initialize DefiLlama MCP toolset
  const toolset = McpDefillama({
    debug: false,
    retryOptions: {
      maxRetries: 3,
      initialDelay: 500,
    },
  });

  // Get available tools
  const defillamaTools = await toolset.getTools();

  // Create agent with DefiLlama tools
  const { runner } = await AgentBuilder.create("defillama_agent")
    .withModel("gemini-2.5-flash")
    .withDescription("An agent that queries DeFi protocol data from DefiLlama")
    .withTools(...defillamaTools)
    .build();

  try {
    // Example queries
    const response = await runner.ask(
      "What are the top DeFi protocols by TVL?",
    );
    console.log(response);
  } finally {
    // Clean up resources
    await toolset.close();
  }
}

main().catch(console.error);

Best Practices

• Resource Cleanup: Always call await toolset.close() when done to properly close connections
• Error Handling: Implement proper error handling for data requests
• Rate Limits: Be aware of API rate limits and implement appropriate retry logic
• LLM Integration: For enhanced resolution, configure an LLM provider via OPENROUTER_API_KEY or GOOGLE_GENERATIVE_AI_API_KEY

Error Handling

🔄 Invalid protocol identifiers: Verify protocol names or IDs are correct 🌐 Network issues: Check your internet connection and DefiLlama API status

Resources

• DefiLlama - DeFi data aggregator platform
• Model Context Protocol - MCP specification and standards