# 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](https://docs.parallel.ai/integrations/mcp/getting-started) for more details.
## Installation
The official installation instructions can be found [here](https://docs.parallel.ai/integrations/mcp/installation).
```json
{
"mcpServers": {
"Parallel Task MCP": {
"url": "https://task-mcp.parallel.ai/mcp"
}
}
}
```
## Running locally
<details><summary>Running locally</summary>
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
</details>
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/janwilmake/task-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server