trackmcp
Back to directory
15 stars TypeScriptOthers Updated Jul 9, 2026

Documentation

Parallel Task MCP

The Parallel Task MCP allows initiating deep research or task groups directly from your favorite LLM client. It can be a great way to get to know Parallel’s different APIs by exploring their capabilities, but can also be used as a way to easily do small experiments while developing production systems using Parallel APIs. Please read our MCP docs here for more details.

Installation

The official installation instructions can be found here.

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

Running locally

Running locally

This repo contains a proxy to the mcp which is hosted at: https://task-mcp.parallel.ai/mcp

How to run and test locally:

1. `wrangler dev`

2. `npx @modelcontextprotocol/inspector`

3. Connect to server: http://localhost:8787/mcp

Frequently asked questions

What is task-mcp?

task-mcp is a Model Context Protocol (MCP) server listed in the TrackMCP directory.

How do I install task-mcp?

Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.

Is task-mcp open source?

Yes — it is hosted on GitHub at https://github.com/parallel-web/task-mcp and has 15 stars.

Related MCP tools

Run your own MCP server? See who uses it and what to fix.

Measure it with TrackMCP