Agentai MCP Server

🚀 Agent.ai MCP Server

An MCP server implementation integrated into the Agent.ai API, offering web text extraction, web screenshot capture, and YouTube subtitle retrieval through a dynamic function loading system.

✨ Features

• Dynamic Function Loading: Automatically fetches available functions from the Agent.ai API.
• Web Text Extraction: Scrapes or crawls web pages to obtain text content.
• Web Screenshot: Captures visual screenshots of web pages.
• YouTube Subtitles: Extracts subtitles from YouTube videos.
• Caching: Efficiently caches function definitions to reduce API calls.

🛠️ Tools

This server dynamically loads tools from the Agent.ai API. Currently available tools include:

• grab_web_text

  • Extracts the text content of a web page.
  • Input:
    • url (string, required): The URL of the web page to extract.
    • mode (string, optional): "Scrape" a single page or "Crawl" up to 100 pages.

• grab_web_screenshot

  • Captures a visual screenshot of a web page.
  • Input:
    • url (string, required): The URL of the web page to capture.
    • ttl_for_screenshot (integer, optional): The expiration time (in seconds) for the screenshot cache.

• get_youtube_transcript

  • Retrieves the subtitles of a YouTube video.
  • Input:
    • url (string, required): The URL of the YouTube video.

And dozens of other tools. To view all available tools, visit https://docs.agent.ai/api-reference.

⚙️ Configuration

Obtain an API Token

Using this MCP server requires an Agent.ai API token. Please contact Agent.ai to get your token.

Use with Claude Desktop

Add the following content to claude_desktop_config.json:

Docker

{
  "mcpServers": {
    "agentai": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "API_TOKEN",
        "mcp/agentai"
      ],
      "env": {
        "API_TOKEN": "YOUR_API_TOKEN_HERE"
      }
    }
  }
}

NPX

{
  "mcpServers": {
    "agentai": {
      "command": "npx",
      "args": [
        "-y",
        "@agentai/mcp-server"
      ],
      "env": {
        "API_TOKEN": "YOUR_API_TOKEN_HERE"
      }
    }
  }
}

💻 Usage Examples

Extract Web Text

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: '{"url":"https://agent.ai","mode":"Scrape"}'
};

fetch('https://api-lr.agent.ai/api/v1/grab_web_text', options)
  .then(response => response.json())
  .then(data => console.log(data))
  .catch(error => console.error('Error:', error));

Capture Web Screenshot

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: '{"url":"https://agent.ai","mode":"Crawl"}'
};

fetch('https://api-lr.agent.ai/api/v1/grab_web_screenshot', options)
  .then(response => response.blob())
  .then(blob => {
    const img = document.createElement('img');
    img.src = URL.createObjectURL(blob);
    document.body.appendChild(img);
  })
  .catch(error => console.error('Error:', error));

Get YouTube Transcript

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: '{"url":"https://www.youtube.com/watch?v=VIDEO_ID"}'
};

fetch('https://api-lr.agent.ai/api/v1/get_youtube_transcript', options)
  .then(response => response.json())
  .then(data => console.log(data))
  .catch(error => console.error('Error:', error));

🏗️ Build

docker build -t mcp/agentai .

📄 License

This software is licensed under [License Name]. Please refer to [LICENSE File Path] for specific details.