> ## Documentation Index
> Fetch the complete documentation index at: https://docs.taskguru.so/llms.txt
> Use this file to discover all available pages before exploring further.

# Connecting Cursor IDE

> Give your Cursor AI Editor native access to your TaskGuru account using the Model Context Protocol (MCP).

# Cursor Editor Integration (MCP)

Cursor supports the Model Context Protocol (MCP) natively, allowing its AI Composer to directly read and manage your software development tasks stored in TaskGuru.

By connecting Cursor, your AI can automatically pull its next development ticket, read the discussion thread, write the code, and then mark the task as complete. TaskGuru functions perfectly as an AI bot task management layer.

<Steps>
  <Step title="Generate Token">
    Generate a new **Personal Access Token** in your TaskGuru account by navigating to **Settings > API Keys**.
  </Step>

  <Step title="Open Cursor MCP Settings">
    Open Cursor, navigate to **Cursor Settings**, and go to the **Features > MCP** tab.
  </Step>

  <Step title="Add the TaskGuru Server">
    Click **+ Add New MCP Server**.

    Fill out the configuration:

    * **Type**: `command`
    * **Name**: `TaskGuru`
    * **Command**: `npx -y @taskguru/mcp-server`
  </Step>

  <Step title="Add Environment Variables">
    In the environment variables section for this new MCP server, securely add your credentials:

    * `SUPABASE_URL`: `https://your-project.supabase.co`
    * `SUPABASE_ANON_KEY`: `your-anon-key`
    * `TASKGURU_API_KEY`: `tg_pat_YOUR_TOKEN_HERE`
  </Step>

  <Step title="Save and Connect">
    Click Save. Cursor will initialize the server and you will have full access to TaskGuru inside your IDE!
  </Step>
</Steps>
