# Amazon Amazon Clients

Source: https://userguide.qualytics.io/using-the-platform/mcp/connecting-external-ai-clients/amazon/

Connect Kiro and Amazon Q Developer to the Qualytics MCP server. Before you start, have a Personal API Token and your Qualytics MCP server URL ready. See [External AI Client Requirements](https://userguide.qualytics.io/using-the-platform/mcp/connecting-external-ai-clients/requirements/).

**Tab: Kiro**

The Kiro IDE and Kiro CLI read the same MCP configuration, so you add the Qualytics server once and use it in both. Kiro CLI is the new name of the Amazon Q Developer CLI.

1. - `~/.kiro/settings/mcp.json` makes Qualytics available in all your workspaces.
   - `.kiro/settings/mcp.json` in a workspace keeps it in that workspace.

   Open the MCP configuration file you want to use:

   In the Kiro IDE, you can open either file from the Kiro panel with the **Open MCP Config** icon, or from the command palette with **Kiro: Open user MCP config (JSON)** or **Kiro: Open workspace MCP config (JSON)**.
2. Add the Qualytics server under `mcpServers`:

   ```json
   {
     "mcpServers": {
       "qualytics": {
         "url": "https://your-qualytics-instance.qualytics.io/api/mcp/"
       }
     }
   }
   ```
3. Save the file. Kiro reconnects its MCP servers on its own. In the IDE, make sure MCP support is turned on in **Settings** (search for "MCP").
4. Kiro registers itself with Qualytics and opens the Qualytics authorization page in your browser. Paste your Qualytics API token and approve the access.
5. In the IDE, check that Qualytics shows as connected in the **MCP servers** tab of the Kiro panel.

Leave out the `oauth` and `headers` properties. Kiro registers its own client with Qualytics, and Qualytics does not accept an API token in a header.

When the authorization expires, the IDE shows a **Re-authenticate** button in the MCP panel. In Kiro CLI, run `/mcp auth`.

**Tab: Amazon Q Developer in the IDE**

> **Warning: End of support**
>
> AWS ends support for the Amazon Q Developer IDE plugins on April 30, 2027, and recommends moving to Kiro. New users should use the **Kiro** tab.

These steps work in every IDE with the Amazon Q Developer plugin, such as VS Code and JetBrains.

1. Open the Amazon Q Developer panel, then the **Chat** panel.
2. Choose the tools icon to open the MCP configuration.
3. Choose **+** to add a server.
4. Select the scope. **Global** makes Qualytics available in all your projects, and **Local** keeps it in the current project.
5. In **Name**, enter `qualytics`.
6. Select `http` as the transport.
7. In **URL**, enter your Qualytics MCP URL: `https://your-qualytics-instance.qualytics.io/api/mcp/`.
8. Leave **Headers - optional** empty, then choose **Save**.
9. Amazon Q opens the Qualytics authorization page in your browser. Paste your Qualytics API token and approve the access.
10. In the tool permissions panel that opens, choose **Ask**, **Always allow**, or **Deny** for each Qualytics tool.

If Amazon Q shows an alert for the server, choose **Fix Configuration** to return to its settings.

> **Info: Official Documentation**
>
> These steps follow the guides for [MCP in Kiro](https://kiro.dev/docs/mcp/configuration/), [MCP in Amazon Q Developer in the IDE](https://docs.aws.amazon.com/amazonq/latest/qdeveloper-ug/mcp-ide.html), and [upgrading from Amazon Q Developer to Kiro](https://docs.aws.amazon.com/amazonq/latest/qdeveloper-ug/upgrade-to-kiro.html). If a menu or command in your version differs, check those pages.

If the connection fails, see [MCP Troubleshooting](https://userguide.qualytics.io/using-the-platform/mcp/troubleshooting/).

*Last updated: November 7, 2026.*