Name: ADK-TS
Author: IQ AI

Convenient wrappers for popular third-party MCP servers from Anthropic and the community

ADK-TS provides convenience wrappers for popular third-party MCP servers, making them as easy to use as IQ AI's built-in servers. These wrappers follow the same simple configuration pattern.

Community Servers

These are wrappers around MCP servers built by Anthropic and the community. You can also connect to any MCP-compliant server using McpToolset directly.

Learn more about integrating external MCPs →

Market Data

Third-party MCP servers for cryptocurrency market data and analytics.

MCP CoinGecko

Access free cryptocurrency market data and analytics via public API

MCP CoinGecko Pro

Access premium cryptocurrency market data with enhanced features

Productivity

MCP servers for workspace and knowledge management tools.

MCP Notion

Read and write Notion pages, databases, and workspace content

Utilities

MCP servers for file operations, memory management, and utility functions.

MCP Filesystem

File system operations (read/write/list) using Anthropic's filesystem server

MCP Memory

Memory and note-taking capabilities for persistent agent knowledge

MCP Playwright

Browser automation and web scraping using Microsoft's Playwright server

MCP Sequential Thinking

Structured step-by-step reasoning for complex problem solving

Quick Start

IQ AI's Third-Party MCP Wrappers follow the same simple pattern as IQ AI servers:

import { McpFilesystem, McpMemory, LlmAgent } from "@iqai/adk";

// Initialize filesystem with allowed directories
const filesystemToolset = McpFilesystem({
  env: {
    ALLOWED_DIRECTORIES: "/workspace,/project/data",
  },
});

// Initialize memory (no config needed)
const memoryToolset = McpMemory();

// Get tools
const [fileTools, memoryTools] = await Promise.all([
  filesystemToolset.getTools(),
  memoryToolset.getTools(),
]);

// Create agent with combined tools
const agent = new LlmAgent({
  name: "smart_assistant",
  description: "An assistant with file system and memory capabilities",
  model: "gemini-2.5-flash",
  tools: [...fileTools, ...memoryTools],
});

// Always cleanup
await Promise.all([filesystemToolset.close(), memoryToolset.close()]);

Common Configuration

All IQ AI's Third-Party MCP Wrappers support these configuration options:

Option	Type	Description
`env`	object	Environment variables (server-specific)
`debug`	boolean	Enable debug logging
`description`	string	Custom description
`retryOptions.maxRetries`	number	Max retry attempts (default: 2)
`retryOptions.initialDelay`	number	Initial retry delay in ms (default: 200)
`samplingHandler`	SamplingHandler	Handler for MCP sampling requests

Using Other External MCPs

No wrapper for the MCP server you need? Use McpGeneric to connect to any npm-based MCP server with the same config pattern as the built-in wrappers:

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

// Connect to any npm-based MCP server
const toolset = McpGeneric("@some-org/mcp-server-package", {
  env: {
    API_KEY: process.env.API_KEY || "",
  },
});

const tools = await toolset.getTools();

You can also pass a custom display name as a third argument:

const toolset = McpGeneric(
  "@modelcontextprotocol/server-brave-search",
  { env: { BRAVE_API_KEY: process.env.BRAVE_API_KEY || "" } },
  "Brave Search MCP",
);

For more control (SSE transport, custom retry logic, etc.), use McpToolset directly:

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

const toolset = new McpToolset({
  name: "Custom MCP Server",
  description: "Any MCP-compliant server",
  transport: {
    mode: "stdio",
    command: "npx",
    args: ["-y", "@some-org/mcp-server-package"],
    env: {
      API_KEY: process.env.API_KEY || "",
      PATH: process.env.PATH || "",
    },
  },
});

const tools = await toolset.getTools();