> For the complete documentation index, see [llms.txt](https://docs.feathery.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.feathery.io/develop/mcp-server/connect-to-the-mcp-server.md).

# Connect to the MCP Server

Connect Claude, ChatGPT, Claude Code, Cursor, VS Code, Codex, and other AI tools to Feathery by signing in with your Feathery account or using an API key.

{% stepper %}
{% step %}

### Copy the MCP URL for Your Region

| Region        | MCP URL                              |
| ------------- | ------------------------------------ |
| US            | `https://api.feathery.io/api/mcp`    |
| CA            | `https://api-ca.feathery.io/api/mcp` |
| EU            | `https://api-eu.feathery.io/api/mcp` |
| JP            | `https://api-jp.feathery.io/api/mcp` |
| AU            | `https://api-au.feathery.io/api/mcp` |
| {% endstep %} |                                      |

{% step %}

### Add Feathery to Your AI Tool

Most AI tools connect by signing in with your Feathery account. If your tool doesn't support signing in, or you want to work with a test environment, open **Use an API key instead**. The examples use the US URL; replace it with your region's URL if needed.

{% tabs %}
{% tab title="Claude" %}

1. In Claude, open **Settings → Connectors** and choose **Add custom connector**.
2. Name it `Feathery`, enter the MCP URL, and add it.
3. Click **Connect** and sign in with your Feathery account.

On Team and Enterprise plans, an owner adds the connector for the organization first. Members then connect their own Feathery accounts.
{% endtab %}

{% tab title="ChatGPT" %}

1. In ChatGPT, open **Settings → Apps** and turn on **Developer mode** under **Advanced settings**.
2. Choose **Create app**, name it `Feathery`, and enter the MCP URL.
3. Set **Authentication** to **OAuth**, click **Create**, and sign in with your Feathery account.

On Business and Enterprise workspaces, an admin may need to add the app first.
{% endtab %}

{% tab title="Claude Code" %}
Run:

```bash
claude mcp add --transport http feathery https://api.feathery.io/api/mcp
```

Then run `/mcp` in Claude Code, select `feathery`, and choose **Authenticate**.

<details>

<summary>Use an API key instead</summary>

```bash
claude mcp add --transport http feathery https://api.feathery.io/api/mcp \
  --header "Authorization: Token $FEATHERY_API_KEY"
```

</details>
{% endtab %}

{% tab title="Cursor" %}
[Install in Cursor](cursor://anysphere.cursor-deeplink/mcp/install?name=feathery\&config=eyJ1cmwiOiJodHRwczovL2FwaS5mZWF0aGVyeS5pby9hcGkvbWNwIn0%3D) (US region), or add this to `~/.cursor/mcp.json`:

```json
{
  "mcpServers": {
    "feathery": {
      "url": "https://api.feathery.io/api/mcp"
    }
  }
}
```

Then sign in with your Feathery account when Cursor prompts you.

<details>

<summary>Use an API key instead</summary>

```json
{
  "mcpServers": {
    "feathery": {
      "url": "https://api.feathery.io/api/mcp",
      "headers": {
        "Authorization": "Token <your Feathery API key>"
      }
    }
  }
}
```

</details>
{% endtab %}

{% tab title="VS Code" %}
[Install in VS Code](https://vscode.dev/redirect/mcp/install?name=feathery\&config=%7B%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fapi.feathery.io%2Fapi%2Fmcp%22%7D) (US region), or add this to `.vscode/mcp.json`:

```json
{
  "servers": {
    "feathery": {
      "type": "http",
      "url": "https://api.feathery.io/api/mcp"
    }
  }
}
```

Then start the server and sign in with your Feathery account when VS Code prompts you.

<details>

<summary>Use an API key instead</summary>

VS Code prompts for the API key when it starts the server.

```json
{
  "inputs": [
    {
      "type": "promptString",
      "id": "feathery-api-key",
      "description": "Feathery API key",
      "password": true
    }
  ],
  "servers": {
    "feathery": {
      "type": "http",
      "url": "https://api.feathery.io/api/mcp",
      "headers": {
        "Authorization": "Token ${input:feathery-api-key}"
      }
    }
  }
}
```

</details>
{% endtab %}

{% tab title="Codex" %}
Run:

```bash
codex mcp add feathery --url https://api.feathery.io/api/mcp
codex mcp login feathery
```

The second command opens your browser to sign in with your Feathery account. The ChatGPT desktop app and the Codex IDE extension share this configuration.

<details>

<summary>Use an API key instead</summary>

Add this to `~/.codex/config.toml`, then restart Codex:

```toml
[mcp_servers.feathery]
url = "https://api.feathery.io/api/mcp"

[mcp_servers.feathery.http_headers]
Authorization = "Token <your Feathery API key>"
```

</details>
{% endtab %}

{% tab title="Other" %}
Add Feathery as a remote (HTTP) MCP server with the MCP URL. If your tool supports MCP sign-in (OAuth), don't add an `Authorization` header; the tool prompts you to sign in.

If your tool doesn't support signing in, add this header:

```
Authorization: Token <your Feathery API key>
```

{% endtab %}
{% endtabs %}

{% hint style="info" %}
To use an API key, copy an environment API key from the **Developer** tab in your Feathery settings. Don't use the SDK key that embeds forms. Use a test environment key when experimenting.
{% endhint %}
{% endstep %}

{% step %}

### Approve Access in Feathery

When you sign in, Feathery shows which app is asking and what it wants to do:

* **Connect to**: the organization the assistant works in. Each connection covers one organization.
* **Work in your workspace**: forms, submissions, users, fields, AI document extractions, templates, and logs. What the assistant can do depends on your role.
* **Read your meetings**: meeting summaries, action items, and transcripts.

Claude and ChatGPT appear as verified apps. Other tools, like Claude Code, Cursor, VS Code, and Codex, may appear as an unverified app; that's expected. For these tools, **Work in your workspace** starts unticked, so tick it if you want the assistant to work with your forms and submissions.

Choose what to allow, click **Allow**, then go back to your AI tool.

{% hint style="info" %}
Connection requests last 10 minutes. If the page says the request expired, start connecting again from your AI tool.
{% endhint %}
{% endstep %}

{% step %}

### Test the Connection

Ask your assistant:

```
List my Feathery forms.
```

If you allowed meetings access, you can also ask:

```
Show my meetings from this week.
```

{% endstep %}
{% endstepper %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.feathery.io/develop/mcp-server/connect-to-the-mcp-server.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.