# Client Setup

Connect Hive's remote MCP server to a supported AI client, verify that the client can discover and call Hive tools, then choose the production surface that fits your app or agent.

## Connection details

| Item | Value |
| --- | --- |
| Remote MCP URL | `https://mcp.hiveintelligence.xyz/mcp` |
| Transport | Streamable HTTP |
| Auth header | `Authorization: Bearer YOUR_HIVE_API_KEY` |
| Key setup | `/dashboard/keys` |

## Before you start

- Verify the official MCP endpoint before approving a client connection: `https://mcp.hiveintelligence.xyz/mcp`. Do not approve a look-alike MCP URL or marketplace entry that points somewhere else.
- Keep your Hive API key private; do not commit it to source.
- Use tool-call approval or review prompts when your client supports them.
- Use an install link when your client supports it, or add the manual configuration for the same client.
- Review `/mcp-security` before enabling stateful tools, B2B subject writes, or downstream actions.

Hive provider tools are read-only data tools. Hive-native stateful tools can write Hive-owned monitors, alerts, memory facts, reports, and B2B subject audit state, so keep those calls scoped to trusted users and backends.

## Claude Desktop

Claude Desktop remote MCP setup belongs in Claude's Custom Connectors UI, not `claude_desktop_config.json`.

```text
Name: Hive
URL: https://mcp.hiveintelligence.xyz/mcp
Authentication: Bearer token
Token: YOUR_HIVE_API_KEY
```

## Claude Code

```bash
claude mcp add --transport http hive https://mcp.hiveintelligence.xyz/mcp \
  --header "Authorization: Bearer YOUR_HIVE_API_KEY"
```

Use `-s user` for a global install or `-s project` when the project should own the MCP entry.

## Cursor

Add Hive through Cursor's MCP settings or `~/.cursor/mcp.json`:

```json
{
  "mcpServers": {
    "hive": {
      "url": "https://mcp.hiveintelligence.xyz/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_HIVE_API_KEY"
      }
    }
  }
}
```

## Windsurf

Add Hive to `~/.codeium/windsurf/mcp_config.json`:

```json
{
  "mcpServers": {
    "hive": {
      "serverUrl": "https://mcp.hiveintelligence.xyz/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_HIVE_API_KEY"
      }
    }
  }
}
```

## VS Code

Create `.vscode/mcp.json` in your project root. VS Code prompts for the Hive key on first use.

```json
{
  "inputs": [
    {
      "type": "promptString",
      "id": "hive-api-key",
      "description": "Hive API Key",
      "password": true
    }
  ],
  "servers": {
    "hive": {
      "type": "http",
      "url": "https://mcp.hiveintelligence.xyz/mcp",
      "headers": {
        "Authorization": "Bearer ${input:hive-api-key}"
      }
    }
  }
}
```

## OpenAI Responses API

Use this from server-side OpenAI API code. ChatGPT developer-mode apps currently document OAuth/no-auth/mixed auth flows, so direct ChatGPT UI setup may require an OAuth-compatible proxy.

```json
{
  "type": "mcp",
  "server_label": "hive",
  "server_description": "Live crypto market, wallet, DeFi, and token-risk tools from Hive Intelligence.",
  "server_url": "https://mcp.hiveintelligence.xyz/mcp",
  "authorization": "YOUR_HIVE_API_KEY",
  "allowed_tools": [
    "search_tools",
    "get_api_endpoint_schema",
    "invoke_api_endpoint"
  ]
}
```

## Codex CLI

Add this block to `~/.codex/config.toml`. For shared machines, prefer an environment variable for the token.

```toml
[mcp_servers.hive]
url = "https://mcp.hiveintelligence.xyz/mcp"
http_headers = { Authorization = "Bearer YOUR_HIVE_API_KEY" }
```

## Gemini CLI

Add Hive to `~/.gemini/settings.json`:

```json
{
  "mcpServers": {
    "hive": {
      "httpUrl": "https://mcp.hiveintelligence.xyz/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_HIVE_API_KEY"
      }
    }
  }
}
```

## CLI

Query Hive from your terminal through the `hive-intelligence` package and `hive` executable:

```bash
HIVE_API_KEY=YOUR_HIVE_API_KEY npx -y -p hive-intelligence@latest hive market price --ids bitcoin --vs usd
```

## REST API

Use REST when your backend, bot, cron job, or serverless function owns execution:

```bash
curl -X POST https://mcp.hiveintelligence.xyz/api/v1/execute \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_HIVE_API_KEY" \
  -d '{"tool":"get_price","args":{"ids":"bitcoin","vs_currencies":"usd"}}'
```

## Verify the connection

1. Save the config in your MCP client.
2. Restart the client or reload its MCP server list.
3. Ask your agent: `"Use Hive tools to get the current Bitcoin price."`

A successful response should come through Hive and include provider-backed market data plus runtime context. Your agent can then use the same connection for market data, DeFi analytics, wallet tracking, and security checks.

## Next steps

- `/get-started` - end-to-end setup path
- `/quick-start` - REST, MCP, category MCP, and CLI examples
- `/install` - per-client install guide index
- `/mcp-security` - prompt-injection boundaries, scoped tool access, and key handling
- `/tools/live-catalog` - current tool catalog and discovery surfaces