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

# Agent Integration

> Setup guides for integrating ModelsLab MCP Web API with Claude Code, OpenCode, Cursor, and other AI coding agents.

Connect ModelsLab's powerful AI generation capabilities to your favorite AI coding assistant. This guide covers integration with popular MCP-compatible agents and tools.

## Overview

The ModelsLab MCP Web API works with any client that supports the [Model Context Protocol](https://modelcontextprotocol.io/) over HTTP. This includes:

* **Claude Code** - Anthropic's CLI coding assistant
* **OpenCode** - Open-source AI coding agent
* **Claude Desktop** - Claude's desktop application
* **Cursor** - AI-powered code editor
* **Continue** - Open-source AI code assistant
* **Custom integrations** - Any MCP-compatible client

## Claude Code

[Claude Code](https://docs.anthropic.com/en/docs/claude-code) is Anthropic's agentic CLI tool for software development.

### Configuration

Add ModelsLab to your Claude Code MCP settings file:

<Tabs>
  <Tab title="macOS / Linux">
    Edit `~/.claude/settings.json`:

    ```json settings.json theme={null}
    {
      "mcpServers": {
        "modelslab": {
          "url": "https://modelslab.com/mcp/v7",
          "headers": {
            "Authorization": "Bearer YOUR_API_KEY"
          }
        }
      }
    }
    ```
  </Tab>

  <Tab title="Windows">
    Edit `%USERPROFILE%\.claude\settings.json`:

    ```json settings.json theme={null}
    {
      "mcpServers": {
        "modelslab": {
          "url": "https://modelslab.com/mcp/v7",
          "headers": {
            "Authorization": "Bearer YOUR_API_KEY"
          }
        }
      }
    }
    ```
  </Tab>
</Tabs>

### Usage

Once configured, you can ask Claude Code to generate images, videos, and audio directly:

```bash theme={null}
claude "Generate an image of a futuristic cityscape at night using ModelsLab"
```

```bash theme={null}
claude "Create a 5-second video of a cat walking using the ModelsLab API"
```

***

## OpenCode

[OpenCode](https://opencode.ai) is an open-source AI coding agent that runs in your terminal.

### Configuration

Add ModelsLab to your OpenCode configuration:

<Tabs>
  <Tab title="opencode.json (Project-level)">
    Create or edit `opencode.json` in your project root:

    ```json opencode.json theme={null}
    {
      "mcp": {
        "servers": {
          "modelslab": {
            "type": "http",
            "url": "https://modelslab.com/mcp/v7",
            "headers": {
              "Authorization": "Bearer YOUR_API_KEY"
            }
          }
        }
      }
    }
    ```
  </Tab>

  <Tab title="Global Config">
    Edit `~/.config/opencode/config.json`:

    ```json config.json theme={null}
    {
      "mcp": {
        "servers": {
          "modelslab": {
            "type": "http",
            "url": "https://modelslab.com/mcp/v7",
            "headers": {
              "Authorization": "Bearer YOUR_API_KEY"
            }
          }
        }
      }
    }
    ```
  </Tab>
</Tabs>

### Usage

Start OpenCode and use ModelsLab tools:

```bash theme={null}
opencode
```

Then in the interactive session:

```
> Generate a logo for my startup using ModelsLab's flux-dev model
```

***

## Claude Desktop

[Claude Desktop](https://claude.ai/download) supports MCP servers for extended functionality.

### Configuration

<Tabs>
  <Tab title="macOS">
    Edit `~/Library/Application Support/Claude/claude_desktop_config.json`:

    ```json claude_desktop_config.json theme={null}
    {
      "mcpServers": {
        "modelslab": {
          "url": "https://modelslab.com/mcp/v7",
          "headers": {
            "Authorization": "Bearer YOUR_API_KEY"
          }
        }
      }
    }
    ```
  </Tab>

  <Tab title="Windows">
    Edit `%APPDATA%\Claude\claude_desktop_config.json`:

    ```json claude_desktop_config.json theme={null}
    {
      "mcpServers": {
        "modelslab": {
          "url": "https://modelslab.com/mcp/v7",
          "headers": {
            "Authorization": "Bearer YOUR_API_KEY"
          }
        }
      }
    }
    ```
  </Tab>
</Tabs>

### Restart Required

After updating the configuration:

1. Completely quit Claude Desktop (not just minimize)
2. Reopen Claude Desktop
3. The ModelsLab tools should now be available

***

## Cursor

[Cursor](https://cursor.sh) is an AI-powered code editor with MCP support.

### Configuration

Add ModelsLab to your Cursor MCP settings:

1. Open Cursor Settings (`Cmd/Ctrl + ,`)
2. Search for "MCP"
3. Add the following server configuration:

```json theme={null}
{
  "mcp.servers": {
    "modelslab": {
      "url": "https://modelslab.com/mcp/v7",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

Alternatively, edit `.cursor/mcp.json` in your project:

```json mcp.json theme={null}
{
  "servers": {
    "modelslab": {
      "type": "http",
      "url": "https://modelslab.com/mcp/v7",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

***

## VS Code (Copilot Agent Mode)

[VS Code](https://code.visualstudio.com/) supports MCP servers through Copilot Agent Mode.

### Configuration

Create or edit `.vscode/mcp.json` in your project:

```json mcp.json theme={null}
{
  "servers": {
    "modelslab-v7": {
      "type": "sse",
      "url": "https://modelslab.com/mcp/v7",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

### Usage

Once configured, use ModelsLab tools in Copilot Agent Mode:

```
@workspace Generate a product image using ModelsLab's text-to-image tool
```

***

## Windsurf

[Windsurf](https://codeium.com/windsurf) is an AI-powered IDE by Codeium with MCP support.

### Configuration

Edit `~/.codeium/windsurf/mcp_config.json`:

```json mcp_config.json theme={null}
{
  "mcpServers": {
    "modelslab-v7": {
      "serverUrl": "https://modelslab.com/mcp/v7",
      "transport": "sse",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

### Restart Required

After updating the configuration, restart Windsurf for the changes to take effect.

***

## Continue

[Continue](https://continue.dev) is an open-source AI code assistant that works with VS Code and JetBrains IDEs.

### Configuration

Edit your Continue configuration file (`~/.continue/config.json`):

```json config.json theme={null}
{
  "mcpServers": [
    {
      "name": "modelslab",
      "transport": {
        "type": "http",
        "url": "https://modelslab.com/mcp/v7",
        "headers": {
          "Authorization": "Bearer YOUR_API_KEY"
        }
      }
    }
  ]
}
```

***

## Generic MCP Client

For any MCP-compatible client, use the following configuration pattern:

```json theme={null}
{
  "servers": {
    "modelslab-v7": {
      "transport": "http",
      "url": "https://modelslab.com/mcp/v7",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

***

## Environment Variables

For security, you can use environment variables instead of hardcoding your API key:

<Tabs>
  <Tab title="macOS / Linux">
    Add to your shell profile (`~/.bashrc`, `~/.zshrc`, etc.):

    ```bash theme={null}
    export MODELSLAB_API_KEY="your_api_key_here"
    ```

    Then reference it in your config:

    ```json theme={null}
    {
      "mcpServers": {
        "modelslab": {
          "url": "https://modelslab.com/mcp/v7",
          "headers": {
            "Authorization": "Bearer ${MODELSLAB_API_KEY}"
          }
        }
      }
    }
    ```
  </Tab>

  <Tab title="Windows">
    Set environment variable:

    ```powershell theme={null}
    setx MODELSLAB_API_KEY "your_api_key_here"
    ```

    Then reference it in your config:

    ```json theme={null}
    {
      "mcpServers": {
        "modelslab": {
          "url": "https://modelslab.com/mcp/v7",
          "headers": {
            "Authorization": "Bearer %MODELSLAB_API_KEY%"
          }
        }
      }
    }
    ```
  </Tab>
</Tabs>

<Warning>
  Note: Environment variable substitution support varies by client. Check your specific client's documentation.
</Warning>

***

## Verification

After configuring your MCP client, verify the integration is working:

### Test Connection

Ask your AI agent to list available models:

```
List the available image generation models from ModelsLab
```

### Generate a Test Image

```
Generate a simple test image of a blue circle on a white background using ModelsLab
```

### Check for Errors

If the integration isn't working, check:

1. **API Key**: Ensure your API key is valid and has sufficient credits
2. **URL**: Verify the endpoint URL is correct
3. **Headers**: Confirm the Authorization header format is correct
4. **Client Support**: Ensure your MCP client supports HTTP transport

***

## Troubleshooting

<AccordionGroup>
  <Accordion title="Connection refused or timeout">
    * Verify the endpoint URL is correct
    * Check your network connection
    * Ensure there are no firewall rules blocking the connection
  </Accordion>

  <Accordion title="Authentication errors">
    * Verify your API key is correct
    * Check that the Authorization header uses `Bearer ` prefix (with space)
    * Ensure your API key has not expired
  </Accordion>

  <Accordion title="Tools not appearing">
    * Restart your MCP client after configuration changes
    * Check the configuration file syntax (valid JSON)
    * Verify the config file is in the correct location
  </Accordion>

  <Accordion title="Rate limiting">
    * Check your ModelsLab account for usage limits
    * Add delays between requests if making many calls
    * Consider upgrading your plan for higher limits
  </Accordion>
</AccordionGroup>

***

## Next Steps

<CardGroup cols={2}>
  <Card title="Tools Reference" icon="wrench" href="/mcp-web-api/tools-reference">
    Explore all available MCP tools and their parameters
  </Card>

  <Card title="Get API Key" icon="key" href="https://modelslab.com/dashboard/api-keys">
    Create or manage your ModelsLab API keys
  </Card>
</CardGroup>