# HTML/CSS to Image API MCP server

> Lets AI agents capture website screenshots, render HTML/CSS, and generate images or PDFs from reusable templates.

- Repository: https://github.com/htmlcsstoimage/agent-plugins
- Page: https://hostedmcp.org/servers/htmlcsstoimage/agent-plugins
- Transport: remote
- Stars: 7
- License: MIT
- Latest version: 1.1.0

## Install

### Claude Code

```sh
claude mcp add --transport http agent-plugins https://mcp.hcti.io/
```

### Cursor (`~/.cursor/mcp.json`)

```json
{
  "mcpServers": {
    "agent-plugins": {
      "url": "https://mcp.hcti.io/"
    }
  }
}
```

### VS Code (`.vscode/mcp.json`)

```json
{
  "servers": {
    "agent-plugins": {
      "type": "http",
      "url": "https://mcp.hcti.io/"
    }
  }
}
```

### Codex (`~/.codex/config.toml`)

```toml
[mcp_servers.agent-plugins]
url = "https://mcp.hcti.io/"
```

### Gemini CLI (`~/.gemini/settings.json`)

```json
{
  "mcpServers": {
    "agent-plugins": {
      "httpUrl": "https://mcp.hcti.io/"
    }
  }
}
```

## FAQ

### How do I authenticate with the HCTI MCP server?

Authentication uses HCTI's browser-based OAuth flow; no API key or access token is stored in this repository.

### Can I generate PDFs with this server?

Yes, the server can render HTML and CSS as either an image or a PDF.

### Does this require managing a local browser?

No, the server allows agents to capture screenshots and render HTML without the need to manage a browser locally.