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

# Quickstart

> Connect your AI client to MedAI Tools in minutes using the hosted server.

**Server URL:** `https://medai-tools.eka.care/mcp`

<Info>
  **Prerequisite:** An active Eka account. On first connect, your AI client will open a browser window for OAuth login. Contact **[ekaconnect@eka.care](mailto:ekaconnect@eka.care)** to get access.
</Info>

Pick your AI client and follow the steps:

<Tabs>
  <Tab title="Claude Desktop">
    ### Claude Desktop (Paid / Pro)

    <Steps>
      <Step title="Open Settings">
        In Claude Desktop, click **Settings** → **Connectors** → **Add custom connector**
      </Step>

      <Step title="Enter server details">
        * Name: `MedAI Tools`
        * URL: `https://medai-tools.eka.care/mcp`
      </Step>

      <Step title="Add and authenticate">
        Click **Add**, then sign in with your Eka account when prompted.
      </Step>
    </Steps>

    ### Claude Desktop (Free)

    Free tier doesn't have the Connectors UI. Use `mcp-remote` as a bridge:

    <Steps>
      <Step title="Open config">
        Go to **Settings** → **Developer** → **Edit Config**
      </Step>

      <Step title="Add this configuration">
        ```json claude_desktop_config.json theme={null}
        {
          "mcpServers": {
            "medai-tools": {
              "command": "npx",
              "args": ["-y", "mcp-remote", "https://medai-tools.eka.care/mcp"]
            }
          }
        }
        ```

        Config location: `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS) or `%APPDATA%\Claude\claude_desktop_config.json` (Windows)
      </Step>

      <Step title="Save and restart">
        Save the file, restart Claude Desktop, then authenticate when the browser opens.
      </Step>
    </Steps>
  </Tab>

  <Tab title="Claude Code">
    Add to your Claude Code settings:

    ```json ~/.claude/settings.json theme={null}
    {
      "mcpServers": {
        "medai-tools": {
          "type": "url",
          "url": "https://medai-tools.eka.care/mcp"
        }
      }
    }
    ```

    Or run from the terminal:

    ```bash theme={null}
    claude mcp add medai-tools --transport http https://medai-tools.eka.care/mcp
    ```
  </Tab>

  <Tab title="ChatGPT">
    **Requirements:** ChatGPT with a paid plan (remote MCP connectors are not available on the Free tier).

    <Steps>
      <Step title="Go to Settings">
        Open [chatgpt.com](https://chatgpt.com) → click your profile → **Settings**
      </Step>

      <Step title="Enable developer mode">
        Navigate to **Apps and Connectors** → toggle on **Developer mode**
      </Step>

      <Step title="Create connector">
        Click **Create app**:

        * Name: `MedAI Tools`
        * URL: `https://medai-tools.eka.care/mcp`
      </Step>

      <Step title="Authenticate">
        Click **Create** and sign in with your Eka account when the OAuth window appears.
      </Step>
    </Steps>
  </Tab>

  <Tab title="Cursor">
    <Steps>
      <Step title="Open MCP settings">
        Press `Cmd/Ctrl + Shift + P` → **Cursor Settings** → **Tools & MCP** → **New MCP Server**
      </Step>

      <Step title="Add configuration">
        ```json theme={null}
        {
          "mcpServers": {
            "medai-tools": {
              "url": "https://medai-tools.eka.care/mcp"
            }
          }
        }
        ```
      </Step>

      <Step title="Connect">
        Click **Connect**, then sign in with your Eka account.
      </Step>
    </Steps>
  </Tab>

  <Tab title="VS Code">
    **Requirements:** VS Code 1.95+ with GitHub Copilot

    <Steps>
      <Step title="Open Command Palette">
        Press `Cmd/Ctrl + Shift + P` → **MCP: Add Server** → **HTTP**
      </Step>

      <Step title="Enter details">
        * URL: `https://medai-tools.eka.care/mcp`
        * Name: `medai-tools`
      </Step>

      <Step title="Authenticate">
        VS Code will prompt you to sign in with your Eka account.
      </Step>
    </Steps>

    Or add directly to settings:

    ```json settings.json theme={null}
    {
      "mcp": {
        "servers": {
          "medai-tools": {
            "type": "http",
            "url": "https://medai-tools.eka.care/mcp"
          }
        }
      }
    }
    ```
  </Tab>
</Tabs>

***

## Verify the Connection

Once connected, test with a prompt in your AI client:

```text theme={null}
Search for Dolo 650 in the Indian drug database
```

You should see the AI call `indian_branded_drug_search` and return drug details including generic composition and manufacturer.

***

## Self-hosted (Partners Only)

The MedAI Tools server source is available to Eka partners. Self-hosting requires Python 3.13+, Docker, AWS Bedrock access, and Eka OAuth credentials.

Contact **[ekaconnect@eka.care](mailto:ekaconnect@eka.care)** to request source access and setup documentation.

***

<Note>
  **Need an Eka account?** Contact **[ekaconnect@eka.care](mailto:ekaconnect@eka.care)** to get started.
</Note>