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

# Connect a client

> Step-by-step setup of the NeuralVerge MCP server in Claude, ChatGPT, Claude Code, Cursor, VS Code, Windsurf, Gemini CLI and MCP Inspector.

# Connect a client

Every client connects to the same URL:

```
https://mcp.neuralverge.ai/mcp
```

Most clients sign you in with OAuth: you add the URL, a browser window opens, you sign in to NeuralVerge and click **Allow**. Clients without browser sign-in use an API key from **Settings → API** in the dashboard instead. Both are described on [MCP Server → Authentication](./mcp-server#authentication).

> [!NOTE]
>
> The server exposes 251 tools. Clients that cap the number of active tools load only part of them — see [Tool count](./mcp-server#tool-count).

## Claude (claude.ai and Claude Desktop)

**1. Add the connector**

Open **Settings → Connectors**, click **Add custom connector**, name it `NeuralVerge` and paste `https://mcp.neuralverge.ai/mcp`. Leave the OAuth fields under **Advanced settings** empty.

On Team and Enterprise plans an Owner adds the connector once under **Organization settings → Connectors**; members then connect it from their own **Settings → Connectors**.

**2. Sign in**

Click **Connect**. Sign in to NeuralVerge in the window that opens and click **Allow**.

**3. Use it in a chat**

In a new chat, enable NeuralVerge from the **+** menu → **Connectors**, then ask a question that needs data.

Connectors added on claude.ai are also available in Claude Desktop and the Claude mobile apps on the same account.

## ChatGPT

Custom connectors require developer mode, available on paid ChatGPT plans.

**1. Turn on developer mode**

Open **Settings → Apps & Connectors → Advanced settings** and enable **Developer mode**.

**2. Create the connector**

In **Settings → Apps & Connectors**, click **Create**. Enter the name `NeuralVerge`, the MCP server URL `https://mcp.neuralverge.ai/mcp`, and choose **OAuth** as the authentication method.

**3. Sign in and use it**

Sign in to NeuralVerge and click **Allow**. In a chat, pick NeuralVerge from the **+** menu → **Developer mode**.

## Claude Code

Add the server, then sign in from inside Claude Code:

```bash theme={null}
claude mcp add --transport http neuralverge https://mcp.neuralverge.ai/mcp
```

Start `claude`, run `/mcp`, select **neuralverge** and choose **Authenticate**. Add `--scope user` to the command to make the server available in every project, or `--scope project` to share it with your team through `.mcp.json`.

For CI or headless use, pass an API key instead of signing in:

```bash theme={null}
claude mcp add --transport http neuralverge https://mcp.neuralverge.ai/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"
```

## Cursor

Add the server to `~/.cursor/mcp.json` (all projects) or `.cursor/mcp.json` (one project):

```json theme={null}
{
  "mcpServers": {
    "neuralverge": {
      "url": "https://mcp.neuralverge.ai/mcp"
    }
  }
}
```

Open **Cursor Settings → MCP**: NeuralVerge shows a **Login** button. Click it, sign in and click **Allow**. To use an API key instead, add `"headers": { "Authorization": "Bearer YOUR_API_KEY" }` next to `url`.

## VS Code (GitHub Copilot)

Add the server to `.vscode/mcp.json` in your workspace, or run **MCP: Add Server** from the Command Palette and choose **HTTP**:

```json theme={null}
{
  "servers": {
    "neuralverge": {
      "type": "http",
      "url": "https://mcp.neuralverge.ai/mcp"
    }
  }
}
```

Click **Start** above the server entry. VS Code asks to sign in; complete the NeuralVerge sign-in in the browser. The tools are available in Copilot Chat in **Agent** mode.

## Windsurf

Open **Windsurf Settings → Cascade → MCP Servers → View raw config** and add the server to `mcp_config.json`. Windsurf connects with an API key:

```json theme={null}
{
  "mcpServers": {
    "neuralverge": {
      "serverUrl": "https://mcp.neuralverge.ai/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

Save and click **Refresh** in the MCP Servers panel.

## Gemini CLI

Add the server to `~/.gemini/settings.json`:

```json theme={null}
{
  "mcpServers": {
    "neuralverge": {
      "httpUrl": "https://mcp.neuralverge.ai/mcp"
    }
  }
}
```

Start `gemini` and run `/mcp auth neuralverge` to sign in. To use an API key instead, add `"headers": { "Authorization": "Bearer YOUR_API_KEY" }`.

## Other clients

Any client that speaks MCP's Streamable HTTP transport works. Point it at `https://mcp.neuralverge.ai/mcp`; if it supports MCP authorization it will offer to sign in, otherwise set the header `Authorization: Bearer YOUR_API_KEY`.

## MCP Inspector

Use the [MCP Inspector](https://github.com/modelcontextprotocol/inspector) to browse and test tools interactively:

```bash theme={null}
npx @modelcontextprotocol/inspector
```

Choose **Streamable HTTP**, enter `https://mcp.neuralverge.ai/mcp`, then either click **Open Auth Settings → Quick OAuth Flow** to sign in, or add the header `Authorization: Bearer YOUR_API_KEY`, and click **Connect**.

## Disconnect a client

Every client that signed in appears under **Settings → API** in the dashboard with a label starting with **Connector:**. Revoke it to cut that client off immediately. Removing the server from the client alone does not revoke its access.

## Troubleshooting

| Symptom | What to check |
| - | - |
| The client can't sign in or "can't register" | Make sure the URL is exactly `https://mcp.neuralverge.ai/mcp`. The older `api.neuralverge.ai/functions/v1/mcp-server` endpoint only accepts API keys. |
| Tools are listed but every call fails with `401` | The connection was revoked, or the API key is wrong. Remove the server from the client and add it again. |
| Calls fail with `402 Payment Required` | Your organization reached its plan or usage limit. See [Errors & Rate Limits](./errors-and-rate-limits). |
| Only some tools are visible | The client caps the number of tools. See [Tool count](./mcp-server#tool-count). |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.