Task MCP

🚀 Parallel Task MCP

The Parallel Task MCP enables users to directly initiate in - depth research or task groups from their preferred LLM client. It serves as an excellent means to explore the capabilities of Parallel's various APIs. Additionally, it can be used for conducting small - scale experiments during the development of production systems that utilize Parallel APIs. For more details, please refer to our MCP docs here.

🚀 Quick Start

The Parallel Task MCP provides a convenient way to interact with Parallel's APIs, allowing for both exploration and experimentation.

✨ Features

• Initiate deep research or task groups from your favorite LLM client.
• Explore the capabilities of Parallel's different APIs.
• Facilitate small - scale experiments during production system development.

📦 Installation

The official installation instructions can be found here.

{
  "mcpServers": {
    "Parallel Task MCP": {
      "url": "https://task-mcp.parallel.ai/mcp"
    }
  }
}

💻 Usage Examples

Basic Usage

If you want to run and test the Parallel Task MCP locally, you can follow these steps:

# Start the local proxy
wrangler dev

# Use the inspector
npx @modelcontextprotocol/inspector

Then, connect to the server at: http://localhost:8787/mcp