Jina AI MCP Server

A Model Context Protocol (MCP) server that provides seamless integration with Jina AI's neural search capabilities. This server enables semantic search, image search, and cross-modal search functionalities through a simple interface.

🚀 Features

Semantic Search: Find semantically similar documents using natural language queries
Image Search: Search for visually similar images using image URLs
Cross-Modal Search: Perform text-to-image or image-to-text searches

📋 Prerequisites

Node.js 16 or higher
A Jina AI account and API key (Get one here)
MCP-compatible environment (e.g., Cline)

🛠️ Installation

Clone the repository:

git clone <repository-url>
cd jina-ai-mcp

Install dependencies:

npm install

Create a .env file with your Jina AI API key:

JINA_API_KEY=your_api_key_here

Build the server:

npm run build

⚙️ Configuration

Add the following configuration to your MCP settings file:

{
  "mcpServers": {
    "jina-ai": {
      "command": "node",
      "args": [
        "/path/to/jina-ai-mcp/build/index.js"
      ],
      "env": {
        "JINA_API_KEY": "your_api_key_here"
      }
    }
  }
}

🔍 Available Tools

1. Semantic Search

Perform semantic/neural search on text documents.

use_mcp_tool({
  server_name: "jina-ai",
  tool_name: "semantic_search",
  arguments: {
    query: "search query text",
    collection: "your-collection-name",
    limit: 10 // optional, defaults to 10
  }
})

2. Image Search

Search for similar images using an image URL.

use_mcp_tool({
  server_name: "jina-ai",
  tool_name: "image_search",
  arguments: {
    imageUrl: "https://example.com/image.jpg",
    collection: "your-collection-name",
    limit: 10 // optional, defaults to 10
  }
})

Perform text-to-image or image-to-text search.

use_mcp_tool({
  server_name: "jina-ai",
  tool_name: "cross_modal_search",
  arguments: {
    query: "a beautiful sunset", // or image URL for image2text
    mode: "text2image", // or "image2text"
    collection: "your-collection-name",
    limit: 10 // optional, defaults to 10
  }
})

📝 Response Format

All search tools return results in the following format:

{
  content: [
    {
      type: "text",
      text: JSON.stringify({
        results: [
          {
            id: string,
            score: number,
            data: Record<string, any>
          }
        ]
      }, null, 2)
    }
  ]
}

🔐 Error Handling

The server handles various error cases:

Invalid API key
Missing or invalid parameters
API rate limits
Network errors
Invalid collection names

All errors are properly formatted and returned with appropriate error codes and messages.

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

Jina AI for their excellent neural search platform
Model Context Protocol for the MCP specification

Jina AI MCP Server

Integrations